Method and system for improving picture clarity in cloud gaming operations

Abstract

The present invention relates to the field of image processing, and discloses a method and system for improving the clarity of images applied to cloud game operations, including: marking the user-intensive areas of target users, establishing edge nodes in the user-intensive areas, and constructing a distributed rendering network of a rendering server cluster; integrating a GANs model and a CNNs model into the distributed rendering network to obtain a clarity processing module, analyzing the target user's picture clarity improvement instruction, and using the clarity processing module to correspond to the GANs model to perform image reconstruction on the game screen to obtain a reconstructed game screen; using the clarity processing module to correspond to the CNNs model to perform resolution processing on the reconstructed game screen to obtain a resolution-processed game screen, and using a preset adaptive flow algorithm to calculate the encoding parameters of the resolution-processed game screen to construct the target user's final game screen. The present invention can improve the clarity improvement effect of the game screen under cloud game operations.

Description

Technical Field

[0001] The present invention relates to a method and system for improving picture clarity applied to cloud game operations, and belongs to the field of image processing. Background Art

[0002] Improving picture clarity under cloud gaming operations means using a series of technical means and optimization strategies to enhance the quality of video streams transmitted to user devices in cloud gaming services, making the game screen more delicate and clear, reducing blur and distortion, and thus improving the user's gaming experience.

[0003] Traditional methods for improving picture clarity in cloud gaming operations mainly involve improving server GPU performance and adopting fixed encoding standards to increase rendering resolution and compress images. However, these methods lack dynamic adaptability and cannot be optimized for different devices, resulting in limited effectiveness in ensuring transmission speed and picture clarity, affecting user experience. Summary of the invention

[0004] The present invention provides a method and system for improving picture clarity applied to cloud game operations, the main purpose of which is to improve the clarity improvement effect of game pictures under cloud game operations.

[0005] To achieve the above purpose, the present invention provides a method for improving picture clarity in cloud gaming operations, comprising:

[0006] Acquire the cloud gaming environment of the target user, analyze the user group needs of the cloud gaming environment, wherein the user group needs include network environment, device performance, and game type preference, and analyze the picture clarity index of the target user according to the user group needs;

[0007] Based on the cloud gaming environment, mark the user-dense area of ​​the target user, establish edge nodes in the user-dense area, establish a rendering server cluster in a preset cloud data center, and construct a distributed rendering network of the rendering server cluster;

[0008] Training the GANs model and the CNNs model of the edge node, integrating the GANs model and the CNNs model into the distributed rendering network, obtaining a definition processing module, and collecting the game screen of the target user based on the edge node;

[0009] Analyze the target user's picture clarity improvement instruction through the game screen and the picture clarity index, map the picture clarity improvement instruction to the target rendering server cluster corresponding to the rendering server cluster, and reconstruct the game screen through the target rendering server cluster using the GANs model corresponding to the clarity processing module to obtain a reconstructed game screen;

[0010] The reconstructed game screen is processed toward resolution using the corresponding CNNs model of the clarity processing module to obtain a resolution-processed game screen, and the encoding parameters of the resolution-processed game screen are calculated using a preset adaptive stream algorithm. The final game screen of the target user is constructed through the encoding parameters.

[0011] Optionally, analyzing user group needs of the cloud gaming environment includes:

[0012] Collecting cloud gaming environment characteristics of the cloud gaming environment;

[0013] Based on the characteristics of the cloud gaming environment, analyzing the network performance, hardware configuration, and proportion of game types of the target users in the cloud gaming environment;

[0014] Analyzing the network environment of the target user through the network performance;

[0015] Analyzing the device performance of the target user according to the hardware configuration;

[0016] Based on the proportion of the game types, determining the game type preference of the target user;

[0017] The user group needs of the target users are analyzed in combination with the network environment, device performance and game type preferences.

[0018] Optionally, analyzing the picture clarity index of the target user according to the user group demand includes:

[0019] Establishing a demand-clarity relationship model between the user group needs and the target user needs;

[0020] Establishing a demand relationship model of the user group's needs;

[0021] Defining demand weights of the user group's demands according to the demand-clarity relationship model;

[0022] Based on the demand-clarity relationship model, the demand relationship model and the demand weight, a picture clarity index of the target user is determined.

[0023] Optionally, marking a user-dense area of ​​the target user based on the cloud gaming environment includes:

[0024] Establishing a geographic coordinate system for the cloud gaming environment;

[0025] Marking the geographic coordinates of the target user in the geographic coordinate system;

[0026] Constructing a heat map of the target user according to the geographic coordinates;

[0027] Calculating the user density of the target user through the heat map;

[0028] Based on the user density, a user-dense area of ​​the target user is determined.

[0029] Optionally, calculating the user density of the target user through the heat map includes:

[0030] Gridding the thermal map to obtain a grid thermal map;

[0031] Mark the number of users in different time intervals on the grid heat map;

[0032] The user density of the target user is calculated using the following formula according to the number of users:

[0033]

[0034] in, represents the user density of target users, Indicates The time weight of the time interval, Indicates In the time interval The number of users in the analysis unit, Represents the grid heat map analysis unit, Represents the grid heat map The analysis unit weight of each analysis unit, Indicates analysis unit, Indicates time interval, Indicates the number of analysis units.

[0035] Optionally, the constructing a distributed rendering network of the rendering server cluster includes:

[0036] Analyzing the rendering load of the rendering server cluster;

[0037] Determining network resources required for the rendering server cluster according to the rendering load;

[0038] Configuring the network switch of the rendering server cluster using the network resources required;

[0039] Calculating network redundancy of the network switch;

[0040] When the network redundancy meets a preset network redundancy threshold, the rendering server cluster is connected to a network to obtain a distributed rendering network of the rendering server cluster.

[0041] Optionally, the reconstructing the game screen by using the target rendering server cluster and using the definition processing module corresponding to the GANs model to obtain the reconstructed game screen includes:

[0042] Extracting screen features of the game screen;

[0043] Determining reconstruction parameters of the game screen according to the screen definition improvement instruction corresponding to the target rendering server cluster and the screen features;

[0044] Based on the reconstruction parameters, using the GANs model to reconstruct the game screen to obtain an initial reconstructed game screen;

[0045] Calculating the image quality of the initial reconstructed game screen;

[0046] When the image quality meets a preset image quality standard, the initial reconstructed game screen is used as the reconstructed game screen.

[0047] Optionally, the calculating the image quality of the initial reconstructed game screen includes:

[0048] Identifying a blurriness metric, a blockiness metric, and a noise metric of the initial reconstructed game screen;

[0049] Analyzing the artifact index of the initial reconstructed game screen according to the blurriness metric, the blockiness metric, and the noise metric;

[0050] Determining a peak signal-to-noise ratio, a structural similarity index, and a perceptual quality metric of the initial reconstructed game picture;

[0051] The image quality of the initial reconstructed game screen is calculated based on the artifact indicator, the peak signal-to-noise ratio, the structural similarity index, and the perceptual quality metric.

[0052] Optionally, the calculating the encoding parameters of the resolution-processed game screen by using a preset adaptive streaming algorithm includes:

[0053] Analyze the screen elements of the game screen at the resolution;

[0054] Determining element encoding requirements of the picture elements;

[0055] Calculating the scene complexity coefficient of the picture element;

[0056] Determine the user network status of the target user corresponding to the resolution-processed game screen;

[0057] The encoding parameters of the resolution-processed game screen are calculated using the adaptive streaming algorithm according to the scene complexity coefficient, the element encoding requirements and the user network status.

[0058] In order to solve the above problems, the present invention also provides a system for improving picture clarity applied to cloud game operations, the system comprising:

[0059] A clarity index determination module is used to obtain the cloud gaming environment of the target user, analyze the user group needs of the cloud gaming environment, wherein the user group needs include network environment, device performance, and game type preference, and analyze the picture clarity index of the target user according to the user group needs;

[0060] A rendering network construction module, used to mark the user-dense area of ​​the target user based on the cloud gaming environment, establish edge nodes in the user-dense area, establish a rendering server cluster in a preset cloud data center, and construct a distributed rendering network of the rendering server cluster;

[0061] A definition processing module constructs a module for training the GANs model and the CNNs model of the edge node, integrating the GANs model and the CNNs model into the distributed rendering network, obtaining a definition processing module, and collecting the game screen of the target user based on the edge node;

[0062] A game screen reconstruction module, used to analyze the target user's screen clarity improvement instruction through the game screen and the screen clarity index, map the screen clarity improvement instruction to a target rendering server cluster corresponding to the rendering server cluster, and reconstruct the game screen through the target rendering server cluster using the GANs model corresponding to the clarity processing module to obtain a reconstructed game screen;

[0063] The game screen encoding module is used to use the clarity processing module corresponding to the CNNs model to perform resolution processing on the reconstructed game screen to obtain a resolution-processed game screen, use a preset adaptive stream algorithm to calculate the encoding parameters of the resolution-processed game screen, and construct the final game screen of the target user through the encoding parameters.

[0064] Compared with the problems described in the background technology, first of all, the cloud gaming environment of the target users and their user group needs are analyzed to ensure the pertinence and effectiveness of the service. Through the comprehensive consideration of the network environment, device performance and game type preference, we can accurately determine the picture clarity index to meet the visual experience needs of different user groups. Secondly, marking the user-dense areas and establishing edge nodes in these areas greatly reduces the transmission delay of the game screen and improves the response speed. The rendering server cluster established in the cloud data center effectively disperses the computing load and improves the rendering efficiency through the distributed rendering network. Furthermore, the integrated application of the GANs model and the CNNs model makes the game screen not only more detailed but also significantly improved in the reconstruction and super-resolution processing process. This image processing technology based on deep learning brings users a high-definition gaming experience that is close to or even beyond local rendering. Finally, the adaptive streaming algorithm is used to calculate the encoding parameters to ensure that users can obtain the best game screen quality under different network conditions. This flexible encoding strategy not only optimizes bandwidth usage, but also reduces the screen freeze and delay problems caused by network fluctuations. Therefore, the present invention can improve the clarity improvement effect of the game screen under cloud game operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of a process for improving picture clarity in cloud gaming operations provided by an embodiment of the present invention;

[0066] Figure 2 A schematic diagram of a module for implementing the method for improving picture clarity applied to cloud gaming operations, provided in accordance with an embodiment of the present invention.

[0067] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0068] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0069] The embodiment of the present application provides a method for improving the picture clarity applied to cloud game operations. The execution subject of the method for improving the picture clarity applied to cloud game operations includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for improving the picture clarity applied to cloud game operations can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0070] Embodiment 1:

[0071] Reference Figure 1 FIG. 1 is a flow chart of a method for improving picture clarity applied to cloud game operations according to an embodiment of the present invention. In this embodiment, the method for improving picture clarity applied to cloud game operations includes:

[0072] S1. Obtain the cloud gaming environment of the target user, analyze the user group needs of the cloud gaming environment, wherein the user group needs include network environment, device performance, and game type preference, and analyze the picture clarity index of the target user according to the user group needs.

[0073] It should be explained that the target users refer to specific game players or user groups, and the cloud gaming environment refers to the specific environment in which these target users conduct gaming activities.

[0074] The present invention analyzes the user group needs of the cloud gaming environment and can systematically analyze the user group needs of the cloud gaming environment, optimize gaming services accordingly, and improve user experience.

[0075] In detail, the analyzing the user group needs of the cloud gaming environment includes:

[0076] Collecting cloud gaming environment characteristics of the cloud gaming environment;

[0077] Based on the characteristics of the cloud gaming environment, analyzing the network performance, hardware configuration, and proportion of game types of the target users in the cloud gaming environment;

[0078] Analyzing the network environment of the target user through the network performance;

[0079] Analyzing the device performance of the target user according to the hardware configuration;

[0080] Based on the proportion of the game types, determining the game type preference of the target user;

[0081] The user group needs of the target users are analyzed in combination with the network environment, device performance and game type preferences.

[0082] Among them, the cloud gaming environment characteristics refer to various environmental factors that affect the gaming experience, including but not limited to network conditions, hardware capabilities, operating system, resolution, game settings, etc.; the network performance refers to indicators such as the speed, stability, delay, and packet loss rate of the user's network connection; the hardware configuration refers to the hardware parameters of the user's device, such as the processor, graphics card, memory, and storage space; the game type ratio refers to the popularity or usage ratio of different game types (such as action, strategy, role-playing, etc.) among the user group; the network environment refers to the specific situation of the network in which the user is playing the game, including Wi-Fi, mobile data, wired connection, etc.; the device performance refers to the comprehensive processing capability of the user's device, the game quality and frame rate that it can support, etc.; the game type preference refers to the user's inclination and preference for different types of games; and the user group demand refers to the expectations and demands of the target users in terms of gaming experience, functions, content, etc.

[0083] The present invention can effectively analyze the picture clarity index of the target user by analyzing the picture clarity index of the target user according to the needs of the user group, and optimize the cloud gaming environment according to the analysis results to improve the user experience.

[0084] In detail, analyzing the picture clarity index of the target user according to the needs of the user group includes:

[0085] Establishing a demand-clarity relationship model between the user group needs and the target user needs;

[0086] Establishing a demand relationship model of the user group's needs;

[0087] Defining demand weights of the user group's demands according to the demand-clarity relationship model;

[0088] Based on the demand-clarity relationship model, the demand relationship model and the demand weight, a picture clarity index of the target user is determined.

[0089] The demand-clarity relationship model refers to the relationship between the specific needs of the user group (such as network environment, device performance, game type preference) and the game screen clarity indicators (such as resolution, frame rate, texture details, etc.), and the demand relationship model refers to the relationship and dependency between different user needs. For example, the demand for network bandwidth may affect the demand for device performance, or the game type preference may affect the user's expectation of picture clarity. The demand weight refers to the importance of each demand factor in different user needs, and the picture clarity indicator refers to a set of standards for measuring the quality of game pictures.

[0090] S2. Based on the cloud gaming environment, mark the user-dense area of ​​the target user, establish edge nodes in the user-dense area, establish a rendering server cluster in a preset cloud data center, and construct a distributed rendering network of the rendering server cluster.

[0091] The present invention is based on the cloud gaming environment, and marking the user-dense area of ​​the target user can effectively mark the target user-dense area in the cloud gaming environment, and provide valuable information for game operators to improve service quality and user experience.

[0092] In detail, marking the user-dense area of ​​the target user based on the cloud gaming environment includes:

[0093] Establishing a geographic coordinate system for the cloud gaming environment;

[0094] Marking the geographic coordinates of the target user in the geographic coordinate system;

[0095] Constructing a heat map of the target user according to the geographic coordinates;

[0096] Calculating the user density of the target user through the heat map;

[0097] Based on the user density, a user-dense area of ​​the target user is determined.

[0098] Among them, the geographic coordinate system refers to a system for determining the position on the surface of the earth, which is usually composed of longitude, latitude and altitude (optional). The geographic coordinates refer to a set of numerical values ​​used to accurately describe the position of a point in a geographic coordinate system. The heat map refers to a data visualization tool that represents the density or frequency of data points by the intensity of color. The user density refers to a measure of the number of users in a specific area. The user-dense area refers to an area with a relatively large number of users in a cloud gaming environment.

[0099] Furthermore, calculating the user density of the target user through the heat map includes:

[0100] Gridding the thermal map to obtain a grid thermal map;

[0101] Mark the number of users in different time intervals on the grid heat map;

[0102] The user density of the target user is calculated using the following formula according to the number of users:

[0103]

[0104] in, represents the user density of target users, Indicates The time weight of the time interval, Indicates In the time interval The number of users in the analysis unit, Represents the grid heat map analysis unit, Represents the grid heat map The analysis unit weight of each analysis unit, Indicates analysis unit, Indicates time interval, Indicates the number of analysis units.

[0105] Among them, the grid heat map refers to a heat map that divides the geographical area of ​​the cloud gaming environment into a series of grid units. The number of users refers to the number of users located in a certain analysis unit on the grid heat map within a specific time interval. The time interval refers to the time period divided when analyzing user density, which can be hours, days, weeks, etc. The time weight refers to the weight coefficient assigned to different time intervals, which is used to reflect the importance of user activities in different time periods. The analysis unit refers to a single grid on the grid heat map, which is the basic unit for calculating user density.

[0106] The present invention establishes edge nodes in user-dense areas to reduce the delay in users accessing game content, increase data transmission speed, reduce network congestion, and thus enhance the user's gaming experience. The edge nodes refer to nodes located at the edge of the network and close to user devices. The rendering server cluster refers to a group of interconnected servers that work together to provide high-performance rendering services.

[0107] The present invention constructs a distributed rendering network of the rendering server cluster to ensure that the cluster can run efficiently and stably.

[0108] In detail, the construction of the distributed rendering network of the rendering server cluster includes:

[0109] Analyzing the rendering load of the rendering server cluster;

[0110] Determining network resources required for the rendering server cluster according to the rendering load;

[0111] Configuring the network switch of the rendering server cluster using the network resources required;

[0112] Calculating network redundancy of the network switch;

[0113] When the network redundancy meets a preset network redundancy threshold, the rendering server cluster is connected to a network to obtain a distributed rendering network of the rendering server cluster.

[0114] Among them, the rendering load refers to the workload undertaken by the rendering server cluster when executing rendering tasks, the network required resources refer to the resources that the network infrastructure of the rendering server cluster must provide in order to support the rendering load, the network switch refers to a device used to forward data packets between various servers and other network devices in the rendering server cluster, the network redundancy refers to the additional capacity and path included in the network design, the network redundancy threshold refers to the minimum standard that the network redundancy must meet in order to maintain high availability and performance of the network, and the distributed rendering network refers to a network composed of multiple rendering servers, which are connected together through a high-speed network to jointly execute rendering tasks.

[0115] Optionally, the calculating of the network redundancy of the network switch may be implemented through failover and failback technologies.

[0116] S3. Train the GANs model and the CNNs model of the edge node, integrate the GANs model and the CNNs model into the distributed rendering network, obtain a clarity processing module, and collect the game screen of the target user based on the edge node.

[0117] It should be explained that the GANs model refers to a model used for image reconstruction in the definition processing process, and the CNNs model refers to a model used for image super-resolution in the definition processing process. In detail, the GANs model and CNNs model of the edge node are trained through a large number of low-resolution and high-resolution image pairs.

[0118] The present invention integrates the GANs model and the CNNs model into the distributed rendering network, and the clarity processing module can form a clarity processing center, thereby providing high-quality game images and enhancing user experience. Among them, the clarity processing module refers to a software component that integrates the GANs (Generative Adversarial Network) model and the CNNs (Convolutional Neural Network) model, which is located on the edge node of the distributed rendering network. The main function of this module is to receive image data from the game rendering process, process it through the GANs and CNNs models to improve the clarity and resolution of the image, and then output the optimized image data. The game screen refers to the image data rendered from the game engine, which represents the visual elements such as scenes, characters, and actions in the game world.

[0119] S4. Analyze the target user's picture clarity improvement instruction through the game screen and the picture clarity index, map the picture clarity improvement instruction to the target rendering server cluster corresponding to the rendering server cluster, and reconstruct the game screen through the target rendering server cluster using the GANs model corresponding to the clarity processing module to obtain a reconstructed game screen.

[0120] It should be explained that the picture clarity improvement instruction refers to an instruction generated by a user or a system, which is used to instruct the rendering system to improve the clarity of a specific game picture, and the target rendering server cluster refers to a group of servers in a distributed rendering network, which are assigned to process a specific user's request to improve picture clarity.

[0121] The present invention uses the target rendering server cluster and the clarity processing module corresponding to the GANs model to reconstruct the game screen, and obtains a reconstructed game screen. The GANs model is used to effectively reconstruct the game screen to provide a higher-definition gaming experience.

[0122] In detail, the target rendering server cluster is used to reconstruct the game screen using the definition processing module corresponding to the GANs model to obtain the reconstructed game screen, including:

[0123] Extracting screen features of the game screen;

[0124] Determining reconstruction parameters of the game screen according to the screen definition improvement instruction corresponding to the target rendering server cluster and the screen features;

[0125] Based on the reconstruction parameters, using the GANs model to reconstruct the game screen to obtain an initial reconstructed game screen;

[0126] Calculating the image quality of the initial reconstructed game screen;

[0127] When the image quality meets a preset image quality standard, the initial reconstructed game screen is used as the reconstructed game screen.

[0128] Among them, the picture features refer to the mathematical representation of visual information extracted from the game screen, and these features can be low-level visual features such as color, texture, shape, edge, brightness, etc. The reconstruction parameters refer to the configuration settings used to guide the GANs model to perform image reconstruction, including upsampling ratio, noise level, color correction coefficient, detail enhancement degree, etc. The initial reconstructed game screen refers to the image obtained after the GANs model reconstructs the original game screen for the first time according to the reconstruction parameters. The image quality refers to the visual quality of the reconstructed game screen. The image quality standard refers to a pre-set image quality assessment threshold or indicator used to determine whether the reconstructed game screen reaches an acceptable level. The reconstructed game screen refers to the final game screen processed by the GANs model and meets the image quality standard.

[0129] Furthermore, the calculating the image quality of the initial reconstructed game screen includes:

[0130] Identifying a blurriness metric, a blockiness metric, and a noise metric of the initial reconstructed game screen;

[0131] Analyzing the artifact index of the initial reconstructed game screen according to the blurriness metric, the blockiness metric, and the noise metric;

[0132] Determining a peak signal-to-noise ratio, a structural similarity index, and a perceptual quality metric of the initial reconstructed game picture;

[0133] Based on the artifact index, the peak signal-to-noise ratio, the structural similarity index, and the perceptual quality metric, the image quality of the initial reconstructed game screen is calculated using the following formula:

[0134]

[0135] in, Indicates the image quality of the initial reconstructed game screen. Indicator representing the artifacts of the initial reconstructed game screen, Represents the peak signal-to-noise ratio of the initial reconstructed game screen, Represents the structural similarity index of the initial reconstructed game screen, represents the perceptual quality metric of the initial reconstructed game screen, Indicates the initial reconstruction of the game screen, Indicates the game screen corresponding to the initial reconstructed game screen, represents the weight of the artificial trace indicator, represents the weight of the peak signal-to-noise ratio, represents the weight of the structural similarity index, Represents the weight of the perceptual quality metric.

[0136] Among them, the blur metric refers to an indicator used to evaluate image clarity, the block effect metric refers to an indicator used to evaluate the degree of block distortion in an image, the noise metric refers to an indicator used to evaluate the intensity of random noise in an image, the artificial trace indicator refers to an indicator used to evaluate various distortion indicators introduced in the image reconstruction process, the peak signal-to-noise ratio refers to an objective indicator used to compare image compression quality, the structural similarity index refers to an indicator for measuring the structural similarity of two images, and the perceptual quality metric refers to an indicator for evaluating image quality based on the characteristics of the human visual system.

[0137] S5. Use the clarity processing module corresponding to the CNNs model to perform resolution processing on the reconstructed game screen to obtain a resolution-processed game screen, use a preset adaptive stream algorithm to calculate encoding parameters of the resolution-processed game screen, and construct the final game screen of the target user through the encoding parameters.

[0138] The present invention utilizes the clarity processing module corresponding to the CNNs model to perform resolution processing on the reconstructed game screen, and the obtained resolution-processed game screen can improve the resolution of the reconstructed game screen, thereby providing users with a clearer and more delicate game visual experience. Among them, the resolution-processed game screen refers to the process result of improving the resolution of the game screen through specific image processing technology (such as super-resolution technology). In detail, the use of the clarity processing module corresponding to the CNNs model to perform resolution processing on the reconstructed game screen is that the model further refines the image details through the convolution layer, reduces blur and jagged distortion, and generates a more natural image appearance.

[0139] The present invention uses a preset adaptive streaming algorithm to calculate the encoding parameters of the resolution-processed game screen to ensure that users can obtain the best game screen experience under different network and device conditions, while maximizing the use of network resources and reducing freezes and delays.

[0140] In detail, the method of calculating the encoding parameters of the resolution-processed game screen using a preset adaptive streaming algorithm includes:

[0141] Analyze the screen elements of the game screen at the resolution;

[0142] Determining element encoding requirements of the picture elements;

[0143] Calculating the scene complexity coefficient of the picture element;

[0144] Determine the user network status of the target user corresponding to the resolution-processed game screen;

[0145] The encoding parameters of the resolution-processed game screen are calculated using the adaptive streaming algorithm according to the scene complexity coefficient, the element encoding requirements and the user network status.

[0146] Among them, the screen elements refer to the various visual components in the game screen, including characters, backgrounds, special effects, texts and UI elements. The element encoding requirements refer to the technical requirements that need to be met when encoding screen elements in order to achieve the expected visual quality and playback performance. The scene complexity coefficient refers to the degree of complexity used to describe the overall game screen. The user network status refers to the network connection status between the user device and the server. The encoding parameters refer to the settings used to guide the video encoding process.

[0147] Optionally, the calculating of the scene complexity coefficient of the picture element may be performed by identifying feature point detection.

[0148] Finally, the present invention constructs the final game screen of the target user through the encoding parameters to improve the clarity of the game screen. The final game screen refers to the final game screen with improved clarity.

[0149] Compared with the problems described in the background technology, first of all, the cloud gaming environment of the target users and their user group needs are analyzed to ensure the pertinence and effectiveness of the service. Through the comprehensive consideration of the network environment, device performance and game type preference, we can accurately determine the picture clarity index to meet the visual experience needs of different user groups. Secondly, marking the user-dense areas and establishing edge nodes in these areas greatly reduces the transmission delay of the game screen and improves the response speed. The rendering server cluster established in the cloud data center effectively disperses the computing load and improves the rendering efficiency through the distributed rendering network. Furthermore, the integrated application of the GANs model and the CNNs model makes the game screen not only more detailed but also significantly improved in the reconstruction and super-resolution processing process. This image processing technology based on deep learning brings users a high-definition gaming experience that is close to or even beyond local rendering. Finally, the adaptive streaming algorithm is used to calculate the encoding parameters to ensure that users can obtain the best game screen quality under different network conditions. This flexible encoding strategy not only optimizes bandwidth usage, but also reduces the screen freeze and delay problems caused by network fluctuations. Therefore, the present invention can improve the clarity improvement effect of the game screen under cloud game operation.

[0150] Embodiment 2:

[0151] like Figure 2 The figure shows a functional module diagram of a system for improving picture clarity applied to cloud gaming operations according to the present invention.

[0152] The picture clarity improvement system 200 applied to cloud game operation described in the present invention can be installed in an electronic device. According to the functions implemented, the picture clarity improvement system applied to cloud game operation can include a clarity index determination module 201, a rendering network construction module 202, a clarity processing module construction module 203, a game screen reconstruction module 204 and a game screen encoding module 205. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0153] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0154] The clarity index determination module 201 is used to obtain the cloud gaming environment of the target user, analyze the user group needs of the cloud gaming environment, wherein the user group needs include network environment, device performance and game type preference, and analyze the picture clarity index of the target user according to the user group needs;

[0155] The rendering network construction module 202 is used to mark the user-dense area of ​​the target user based on the cloud game environment, establish the edge node of the user-dense area, establish a rendering server cluster in a preset cloud data center, and construct a distributed rendering network of the rendering server cluster;

[0156] The definition processing module construction module 203 is used to train the GANs model and the CNNs model of the edge node, integrate the GANs model and the CNNs model into the distributed rendering network, obtain the definition processing module, and collect the game screen of the target user based on the edge node;

[0157] The game screen reconstruction module 204 is used to analyze the target user's screen clarity improvement instruction through the game screen and the screen clarity index, map the screen clarity improvement instruction to the target rendering server cluster corresponding to the rendering server cluster, and reconstruct the game screen through the target rendering server cluster using the GANs model corresponding to the clarity processing module to obtain a reconstructed game screen;

[0158] The game screen encoding module 205 is used to use the CNNs model corresponding to the clarity processing module to perform resolution processing on the reconstructed game screen to obtain a resolution-processed game screen, use a preset adaptive stream algorithm to calculate the encoding parameters of the resolution-processed game screen, and construct the final game screen of the target user through the encoding parameters.

[0159] In detail, each module in the picture clarity improvement system 200 applied to cloud game operation in the embodiment of the present invention is used in the same manner as described above. Figure 1 The same technical means as described in the method for improving picture clarity under cloud game operation are used and can produce the same technical effects, so they will not be repeated here.

[0160] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0161] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for improving picture clarity in cloud gaming operations, characterized in that: The method comprises: Acquire the cloud gaming environment of the target user, analyze the user group needs of the cloud gaming environment, wherein the user group needs include network environment, device performance, and game type preference, and analyze the picture clarity index of the target user according to the user group needs; Based on the cloud gaming environment, mark the user-dense area of ​​the target user, establish edge nodes in the user-dense area, establish a rendering server cluster in a preset cloud data center, and construct a distributed rendering network of the rendering server cluster; Training the GANs model and the CNNs model of the edge node, integrating the GANs model and the CNNs model into the distributed rendering network, obtaining a definition processing module, and collecting the game screen of the target user based on the edge node; Analyze the target user's picture clarity improvement instruction through the game screen and the picture clarity index, map the picture clarity improvement instruction to the target rendering server cluster corresponding to the rendering server cluster, and reconstruct the game screen through the target rendering server cluster using the GANs model corresponding to the clarity processing module to obtain a reconstructed game screen; The clarity processing module corresponds to the CNNs model and performs super-resolution processing on the reconstructed game screen to obtain a resolution-processed game screen. The encoding parameters of the resolution-processed game screen are calculated using a preset adaptive stream algorithm. The final game screen of the target user is constructed through the encoding parameters.

2. The method for improving picture clarity in cloud gaming operations as claimed in claim 1, characterized in that: The analyzing user group needs of the cloud gaming environment includes: Collecting cloud gaming environment characteristics of the cloud gaming environment; Based on the characteristics of the cloud gaming environment, analyzing the network performance, hardware configuration, and proportion of game types of the target users in the cloud gaming environment; Analyzing the network environment of the target user through the network performance; Analyzing the device performance of the target user according to the hardware configuration; Based on the proportion of the game types, determining the game type preference of the target user; The user group needs of the target users are analyzed in combination with the network environment, device performance and game type preferences.

3. The method for improving picture clarity in cloud gaming operations as claimed in claim 2, characterized in that: Analyzing the picture clarity index of the target user according to the needs of the user group includes: Establishing a demand-clarity relationship model between the user group needs and the target user needs; Establishing a demand relationship model of the user group's needs; Defining demand weights of the user group's demands according to the demand-clarity relationship model; Based on the demand-clarity relationship model, the demand relationship model and the demand weight, a picture clarity index of the target user is determined.

4. The method for improving picture clarity in cloud gaming operations as claimed in claim 3, characterized in that: The marking a user-dense area of ​​the target user based on the cloud gaming environment includes: Establishing a geographic coordinate system for the cloud gaming environment; Marking the geographic coordinates of the target user in the geographic coordinate system; Constructing a heat map of the target user according to the geographic coordinates; Calculating the user density of the target user through the heat map; Based on the user density, a user-dense area of ​​the target user is determined.

5. The method for improving picture clarity in cloud gaming operations as claimed in claim 4, characterized in that: The calculating the user density of the target user through the heat map includes: Gridding the thermal map to obtain a grid thermal map; Mark the number of users in different time intervals on the grid heat map; The user density of the target user is calculated using the following formula according to the number of users: in, represents the user density of target users, Indicates The time weight of the time interval, Indicates In the time interval The number of users in the analysis unit, Represents the grid heat map analysis unit, Represents the grid heat map The analysis unit weight of each analysis unit, Indicates analysis unit, Indicates time interval, Indicates the number of analysis units.

6. The method for improving picture clarity in cloud gaming operations as claimed in claim 5, characterized in that: The step of constructing the distributed rendering network of the rendering server cluster includes: Analyzing the rendering load of the rendering server cluster; Determining network resources required for the rendering server cluster according to the rendering load; Configuring the network switch of the rendering server cluster using the network resources required; Calculating network redundancy of the network switch; When the network redundancy meets a preset network redundancy threshold, the rendering server cluster is connected to a network to obtain a distributed rendering network of the rendering server cluster.

7. The method for improving picture clarity in cloud gaming operations as claimed in claim 6, characterized in that: The process of reconstructing the game screen by using the target rendering server cluster and using the definition processing module corresponding to the GANs model to obtain the reconstructed game screen includes: Extracting screen features of the game screen; Determining reconstruction parameters of the game screen according to the screen definition improvement instruction corresponding to the target rendering server cluster and the screen features; Based on the reconstruction parameters, using the GANs model to reconstruct the game screen to obtain an initial reconstructed game screen; Calculating the image quality of the initial reconstructed game screen; When the image quality meets a preset image quality standard, the initial reconstructed game screen is used as the reconstructed game screen.

8. The method for improving picture clarity in cloud gaming operations as claimed in claim 7, characterized in that: The calculating the image quality of the initial reconstructed game screen includes: Identifying a blurriness metric, a blockiness metric, and a noise metric of the initial reconstructed game screen; Analyzing the artifact index of the initial reconstructed game screen according to the blurriness metric, the blockiness metric, and the noise metric; Determining a peak signal-to-noise ratio, a structural similarity index, and a perceptual quality metric of the initial reconstructed game picture; The image quality of the initial reconstructed game screen is calculated based on the artifact indicator, the peak signal-to-noise ratio, the structural similarity index, and the perceptual quality metric.

9. The method for improving picture clarity in cloud gaming operations as claimed in claim 8, characterized in that: The method of calculating the encoding parameters of the resolution-processed game screen by using a preset adaptive streaming algorithm includes: Analyze the screen elements of the game screen at the resolution; Determining element encoding requirements of the picture elements; Calculating the scene complexity coefficient of the picture element; Determine the user network status of the target user corresponding to the resolution-processed game screen; The encoding parameters of the resolution-processed game screen are calculated using the adaptive streaming algorithm according to the scene complexity coefficient, the element encoding requirements and the user network status.

10. A system for improving picture clarity applied to cloud gaming operations, characterized in that: The system comprises: A clarity index determination module is used to obtain the cloud gaming environment of the target user, analyze the user group needs of the cloud gaming environment, wherein the user group needs include network environment, device performance, and game type preference, and analyze the picture clarity index of the target user according to the user group needs; A rendering network construction module, used to mark the user-dense area of ​​the target user based on the cloud gaming environment, establish edge nodes in the user-dense area, establish a rendering server cluster in a preset cloud data center, and construct a distributed rendering network of the rendering server cluster; A definition processing module constructs a module for training the GANs model and the CNNs model of the edge node, integrating the GANs model and the CNNs model into the distributed rendering network, obtaining a definition processing module, and collecting the game screen of the target user based on the edge node; A game screen reconstruction module, used to analyze the target user's screen clarity improvement instruction through the game screen and the screen clarity index, map the screen clarity improvement instruction to a target rendering server cluster corresponding to the rendering server cluster, and reconstruct the game screen through the target rendering server cluster using the GANs model corresponding to the clarity processing module to obtain a reconstructed game screen; The game screen encoding module is used to use the clarity processing module corresponding to the CNNs model to perform super-resolution processing on the reconstructed game screen to obtain a resolution-processed game screen, use a preset adaptive stream algorithm to calculate the encoding parameters of the resolution-processed game screen, and construct the final game screen of the target user through the encoding parameters.

Images