WO2020108085A1 - Resource allocation method and apparatus, terminal, and storage medium - Google Patents

Abstract

Embodiments of the present application disclose a resource allocation method and apparatus, a terminal, and a storage medium, pertaining to the field of application optimization. The method comprises: when a game application is run, displaying a game screen; taking a screenshot of the game screen to acquire a game screenshot; using a game scene identification model to identify the game screenshot, and determining a game scene corresponding to the game screen, wherein the game scene is used to indicate a running stage of the game application; determining a resource allocation policy corresponding to the game scene; and allocating system resources to the game application according to the resource allocation policy. In the resource allocation method provided by the embodiments of the present application, a terminal operating system can learn to identify a current game scene by means of intelligent analysis without the game application needing to access the terminal operating system, thereby eliminating the need to establish a data channel between the game application and the operating system, and improving the security of the operating system.

Description

Resource configuration method, device, terminal and storage medium

This application requires the priority of the Chinese patent application with the application number 201811457185.3 and the invention titled "Resource Allocation Method, Device, Terminal, and Storage Medium" filed on November 30, 2018, the entire content of which is incorporated by reference in this application .

Technical field

The embodiments of the present application relate to the technical field of application optimization, and in particular, to a resource configuration method, device, terminal, and storage medium.

Background technique

With the continuous development of mobile terminal technology, there are more and more types of application programs in mobile terminals. For example, a mobile application, a social application, a video playback application, an instant communication application, and a shopping application are installed in the mobile terminal.

In order to improve the application experience, mobile terminal manufacturers are constantly improving their hardware configuration to improve the running speed and quality of applications. For example, mobile terminals are configured with high-performance Central Processing Unit (CPU) and Graphics Processing Unit (GPU) to improve the running speed and picture quality of game applications.

Summary of the invention

Embodiments of the present application provide a resource configuration method, device, terminal, and storage medium. The technical solution is as follows:

In one aspect, a resource configuration method is provided, the method including:

When running the game application, display the game screen;

Take a screenshot of the game screen to obtain a screenshot of the game;

Identifying the game screenshot through the game scene recognition model to determine the game scene corresponding to the game screen, where the game scene is used to indicate the running stage of the game application;

Determine a resource allocation strategy corresponding to the game scene;

Allocate system resources to the game application according to the resource configuration strategy.

In another aspect, a resource configuration device is provided, the device including:

The display module is used to display the game screen when running the game application;

A screenshot module, used to take a screenshot of the game screen to obtain a screenshot of the game;

A scene recognition module, used to identify the game screenshot through the game scene recognition model, and determine the game scene corresponding to the game screen, where the game scene is used to indicate the running stage of the game application;

A strategy determination module for determining a resource configuration strategy corresponding to the game scene;

The resource allocation module is configured to allocate system resources to the game application according to the resource configuration strategy.

On the other hand, a terminal is provided. The terminal includes a processor and a memory; the memory stores at least one instruction, and the at least one instruction is used to be executed by the processor to implement the resource as described above Configuration method.

On the other hand, a computer-readable storage medium is provided, where the storage medium stores at least one instruction, and the at least one instruction is used to be executed by a processor to implement the resource configuration method as described above.

BRIEF DESCRIPTION

FIG. 1 is a schematic structural diagram of a terminal provided by an exemplary embodiment of the present application;

2 is a schematic diagram of the implementation of the communication process between the application program and the operating system in the terminal;

3 is a schematic structural diagram of a terminal provided by an exemplary embodiment of the present application;

4 and 5 are schematic diagrams of the communication process between the application program and the operating system in the terminal shown in FIG. 3;

6 is a schematic structural diagram of a terminal provided by another exemplary embodiment of the present application;

7 shows a flowchart of a resource configuration method shown in an exemplary embodiment of the present application;

FIG. 8 shows a flowchart of a resource configuration method shown in another exemplary embodiment of the present application;

9 shows a flowchart of a resource configuration method shown in another exemplary embodiment of the present application;

10 shows a flowchart of a resource configuration method shown in another exemplary embodiment of the present application;

11 is a schematic diagram of an interface for displaying acceleration controls on the upper layer of the game screen;

FIG. 12 shows a structural block diagram of a resource configuration device provided by an embodiment of the present application.

detailed description

To make the objectives, technical solutions, and advantages of the present invention clearer, the following describes the embodiments of the present invention in further detail with reference to the accompanying drawings.

The "plurality" referred to herein refers to two or more. "And/or" describes the relationship of the related objects, indicating that there can be three relationships, for example, A and/or B, which can indicate: there are three conditions: A exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the related object is a "or" relationship.

Please refer to FIG. 1, which shows a structural block diagram of a terminal 100 provided by an exemplary embodiment of the present application. The terminal 100 may be an electronic device capable of running a game application program, such as a smart phone, a tablet computer, a portable personal computer, or the like. The terminal 100 in this application may include one or more of the following components: a processor 110, a memory 120, and an input/output device 130.

The processor 110 may include one or more processing cores. The processor 110 uses various interfaces and lines to connect various parts of the entire terminal 100, and executes the terminal by running or executing instructions, programs, code sets or instruction sets stored in the memory 120, and calling data stored in the memory 120 100 various functions and processing data. Alternatively, the processor 110 may use at least one of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). Various hardware forms. The processor 110 may integrate one or a combination of a central processing unit (Central Processing Unit, CPU), an image processing unit (Graphics Processing Unit, GPU), and a modem. Among them, CPU mainly deals with operating system, user interface and application program, etc.; GPU is used for rendering and rendering of display content; modem is used for handling wireless communication. It can be understood that the above-mentioned modem may not be integrated into the processor 110, and may be implemented by a communication chip alone.

Optionally, the processor 110 may also include an AI (Artificial Intelligence, Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

The memory 120 may include random access memory (RAM) or read-only memory (Read-Only Memory). Optionally, the memory 120 includes a non-transitory computer-readable storage medium. The memory 120 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 120 may include a storage program area and a storage data area, where the storage program area may store instructions for implementing an operating system and instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.) , Instructions for implementing the following method embodiments, the operating system may be an Android system (including a system based on the deep development of the Android system), an IOS system developed by Apple (including a system based on the deep development of the IOS system) ) Or other systems. The storage data area may also store data created by the terminal 100 in use (such as a phone book, audio and video data, and chat history data).

The input/output device 130 may include a touch display screen for receiving a user's touch operation on or near any suitable object using a finger, a touch pen, or the like, and a user interface displaying various application programs. The touch display screen is usually provided on the front panel of the terminal 100. The touch screen can be designed as a full screen, curved screen, or profile screen. The touch display screen can also be designed as a combination of a full screen and a curved screen, and a combination of a special-shaped screen and a curved screen, which is not limited in the embodiments of the present application.

In addition, those skilled in the art may understand that the structure of the terminal 100 shown in the above drawings does not constitute a limitation on the terminal 100, and the terminal may include more or fewer components than the illustration, or a combination of certain Components, or different component arrangements. For example, the terminal 100 further includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, a wireless fidelity (WiFi) module, a power supply, and a Bluetooth module, which will not be repeated here.

The memory 120 can be divided into an operating system space and a user space. The operating system runs in the operating system space, and native and third-party application programs run in the user space. In order to ensure that different third-party applications can achieve better operating results, the operating system allocates corresponding system resources for different third-party applications. However, the requirements of different application scenarios for system resources in the same third-party application are also different. For example, in the local resource loading scenario, the third-party application has higher requirements for disk read speed; in the animation rendering scenario, the first Three-party applications have higher requirements for GPU performance. However, the operating system and the third-party application are independent of each other. The operating system often fails to perceive the current application scenario of the third-party application in time, so that the operating system cannot perform targeted system resource adaptation according to the specific application scenario of the third-party application.

In a possible implementation manner, as shown in FIG. 2, in order for the operating system to distinguish the specific application scenario of the third-party application, it is necessary to get through the data communication between the third-party application and the operating system, so that the operating system can always Obtain the current scene information of the third-party application, and then perform targeted system resource adaptation based on the current scene.

Taking the Android operating system as an example, the programs and data stored in the memory 120 are shown in FIG. 3, and the memory 120 may store a Linux kernel layer 220, a system runtime library layer 240, an application framework layer 260, and an application layer 280, of which The Linux kernel layer 220, the system runtime library layer 240, and the application framework layer 260 belong to the operating system space, and the application layer 280 belongs to the user space. The Linux kernel layer 220 provides low-level drivers for various hardware of the terminal 100, such as display drivers, audio drivers, camera drivers, Bluetooth drivers, Wi-Fi drivers, and power management. The system runtime library layer 240 provides major feature support for the Android system through some C/C++ libraries. For example, the SQLite library provides database support, the OpenGL/ES library provides 3D drawing support, and the Webkit library provides browser kernel support. An Android runtime library (Android Runtime) is also provided in the system runtime library layer 240, which mainly provides some core libraries that can allow developers to write Android applications using the Java language. The application framework layer 260 provides various APIs that may be used when building applications, and developers can also use these APIs to build their own applications, such as activity management, window management, view management, notification management, content providers, Package management, call management, resource management, positioning management. At least one application program runs in the application layer 280, and these application programs may be native application programs built in the operating system, such as a contact program, an SMS program, a clock program, a camera application, etc.; or may be developed by a third-party developer Third-party applications, such as game applications, instant messaging programs, photo beautification programs, shopping programs, etc.

A feasible communication method between the operating system and the third-party application program is shown in FIG. 4, and a software development kit (Software Development Kit, SDK) for communicating with the operating system is embedded in the third-party application program.

Among them, the SDK contains a number of abstract application programming interfaces (Application Programming Interface), and is provided by the operating system developer to third-party application developers, and the third-party application developers embed the SDK into the Tripartite application. After such third-party applications are installed and run on the operating system, they can call the API provided by the SDK to communicate with the operating system.

As shown in FIG. 4, the system runtime library layer 240 may additionally include an interface communication system 242. The interface communication system 242 can be regarded as a subsystem in the operating system, or as an application program embedded in the operating system. The interface communication system 242 is provided with an SDK interface, and a third-party application program calls the API of the embedded SDK and the SDK interface to perform data communication through a binder. In this way, data related to the application scenarios of third-party applications can be transferred to the operating system through the SDK. With the help of the embedded SDK, the operating system can also actively transfer data to third-party applications, or two-way data transmission can be performed between the operating system and third-party applications.

In another feasible communication method, as shown in FIG. 5, the third-party application program can also establish a long connection with the Socket interface of the interface communication system 242 by using the Socket method. The application scenarios of the third-party application program are related. The data can be transferred to the operating system through the long connection.

As shown in FIGS. 4 and 5, different strategy modules can be set in the interface communication system 242. After receiving the data sent by the third-party application, the interface communication system 242 uses the strategy module corresponding to the third-party application to analyze the data To get the corresponding resource adaptation optimization strategy. Based on the resource adaptation optimization strategy obtained by the analysis, the interface communication system 242 notifies the Linux kernel layer 220 through the control interface to perform system resource adaptation optimization. Among them, the control interface can communicate with the Linux kernel layer 220 in the manner of Sysfs.

Optionally, different policy modules in the interface communication system 242 can correspond to different third-party applications (that is, set policy modules for different applications), or different policy modules correspond to different types of third-party applications (that is, for Different types of applications set policy modules), or different policy modules correspond to different system resources (that is, set policy modules for different system resources), or different policy modules correspond to different application scenarios (that is, for different applications) Scenario setting strategy module), the embodiments of the present application do not limit the specific setting mode of the strategy module. .

Among them, the interface communication system 242 can also communicate with the application framework layer 260 through the way of Binder to receive the foreground application information sent by the application framework layer 260, so that based on the foreground application information, only the third-party applications currently running in the foreground System resource optimization.

Taking the operating system as an IOS system as an example, the programs and data stored in the memory 120 are shown in FIG. 6. The IOS system includes: a core operating system layer 320 (Core OS layer), a core service layer 340 (Core Services layer), and a media layer 360 (Media Layer), touchable layer 380 (Cocoa Touch Layer). The core operating system layer 320 includes an operating system kernel, drivers, and low-level program frameworks. These low-level program frameworks provide functions closer to the hardware for use by the program framework at the core service layer 340. The core service layer 340 provides system services and/or program frameworks required by applications, such as a foundation framework, account framework, advertising framework, data storage framework, network connection framework, geographic location framework, sports framework, and so on. The media layer 360 provides audiovisual interfaces for applications, such as graphics and image related interfaces, audio technology related interfaces, video technology related interfaces, and audio and video transmission technology wireless playback (AirPlay) interfaces. The touchable layer 380 provides various commonly used interface-related frameworks for application development. The touchable layer 380 is responsible for user's touch interaction operations on the terminal 100. Such as local notification service, remote push service, advertising framework, game tool framework, message user interface (User Interface, UI) framework, user interface UIKit framework, map framework, etc.

In the framework shown in FIG. 6, frameworks related to most applications include, but are not limited to: the basic framework in the core service layer 340 and the UIKit framework in the touchable layer 380. The basic framework provides many basic object classes and data types to provide the most basic system services for all applications, regardless of the UI. The class provided by the UIKit framework is a basic UI class library for creating touch-based user interfaces. iOS applications can provide UI based on the UIKit framework, so it provides the basic architecture of the application for building user interfaces and drawing. , Handling and user interaction events, responding to gestures, etc.

Among them, the method and principle of implementing data communication between the third-party application program and the operating system in the IOS system can refer to the Android system, and will not be repeated here.

However, when the communication between third-party applications and the operating system is achieved by adapting system resources based on application scenarios, the third-party applications need to be adjusted during the development process, such as those provided by developers of embedded operating systems. SDK; and, in order to prevent the data interface of the operating system from being exposed to illegal applications and cause potential security risks to the system, the operating system only supports data communication with some third-party applications, resulting in a narrower scope of system resource adaptation.

In the resource configuration method provided in the embodiments of the present application, the operating system takes screenshots of the running game application and uses machine learning to identify game screenshots through a pre-trained game scene recognition model to determine the current game The game scene corresponding to the screen, and further allocate system resources to the game application according to the resource configuration strategy corresponding to the game scene. Since there is no need for data communication between the operating system and the game application during the entire system resource allocation process, there is no need to embed the SDK during the development of the game application, and the data interface of the operating system does not need to be exposed to third-party applications, thereby improving Operating system security. The following uses an exemplary embodiment for description.

Please refer to FIG. 7, which shows a flowchart of a resource configuration method according to an exemplary embodiment of the present application. This embodiment is exemplified by the method applied to the terminal 100 shown in FIG. 1. The method includes:

Step 701, when the game application is running, the game screen is displayed.

When the game application is in the foreground running state, the terminal displays a real-time game screen, which constantly changes during the game running process. For example, in the game startup phase, the game screen is the game startup screen; after the game startup is completed, the game screen is the game main interface; when the game is in progress, the game screen is the game progress screen.

Step 702: Take a screenshot of the game screen to obtain a screenshot of the game.

Since the game screen has different characteristics under different game scenes, the terminal in the embodiment of the present application recognizes the current game scene based on the game screen.

In a possible implementation manner, after the game application is started, the terminal operating system takes screenshots of the game screen at predetermined intervals to obtain corresponding game screenshots.

In another possible implementation manner, after the game application is started, the terminal displays the acceleration control above the game screen, and takes a screenshot when the trigger operation of the acceleration control is received.

It should be noted that when the operating system takes a screenshot of the game screen, it directly extracts the game layer through the background to obtain a screenshot of the game. There is no need for the user to manually take a screenshot, and the user will not perceive it in the foreground.

Optionally, screenshots of the game captured by the operating system are cached in memory and deleted directly after scene recognition.

Step 703: Identify the game screenshot through the game scene recognition model to determine the game scene corresponding to the game screen. The game scene is used to indicate the running stage of the game application.

In the embodiment of the present application, a game scene recognition model for identifying a game scene is pre-stored in the terminal. The game scene recognition model may be a convolutional neural network model, a deep learning model, or other models for image recognition.

For the training method of the game scene recognition model, in a possible implementation manner, the game scene recognition model is obtained by training based on a sample game screenshot carrying a game scene identifier.

Optionally, the terminal operating system inputs the game screenshot into the game scene recognition model to obtain several candidate game scenes output by the game scene model and corresponding probabilities, and then determines the candidate game scene with the highest probability as the game scene corresponding to the game screen.

Optionally, the game scene includes at least one of the following: a game startup scene, a game loading scene, a game main interface scene, and a game progress scene.

Illustratively, after the operating system inputs the game screenshot into the game scene recognition model, the game scene output by the game scene recognition model is obtained as the "game progress scene".

Step 704: Determine the resource configuration strategy corresponding to the game scene.

The terminal operating system further determines the resource configuration strategy corresponding to the game scenario according to the game scenario where the game application is currently located. Optionally, the resource configuration strategy includes configuration information of various system resources in the game scenario. It includes at least one of CPU resources, GPU resources, disk resources, memory resources, and network resources. Correspondingly, the configuration information corresponding to the CPU resources can include the CPU operating frequency and the number of CPU cores turned on; the configuration information corresponding to the GPU resources can include the GPU operating frequency and the number of GPU cores turned on; the configuration information corresponding to the disk resources can include disk rotation speed and disk reading Write speed; configuration information corresponding to memory resources can include memory frequency and memory read and write speed; configuration information corresponding to network resources can include network speed and delay.

In a possible implementation manner, the terminal stores the correspondence between the game scene and the resource configuration strategy, and the terminal operating system determines the resource configuration strategy corresponding to the current game scene based on the correspondence. Schematically, the corresponding relationship is shown in Table 1.

Table I

Game scene

Resource allocation strategy

Game loading scene	Resource Allocation Strategy A
Game main interface scene	Resource Allocation Strategy B
Game progress scene	Resource Allocation Strategy C

Illustratively, when the identified game scene is a "game progress scene", the terminal operating system determines that the resource configuration strategy is "resource configuration strategy C" based on Table 1.

Step 705: Assign system resources to the game application according to the resource configuration strategy.

Optionally, according to the configuration information corresponding to each system resource in the resource configuration strategy, the terminal operating system allocates system resources to the game application program.

In a possible implementation, when the target system resource configuration indicated by the resource configuration strategy is lower than the current system resource configuration, the operating system down-regulates the current system resource configuration to the target system resource configuration indicated by the resource configuration strategy, thereby ensuring the game At the same time as the running quality of the application, it reduces the power consumption of the terminal; when the target system resource configuration indicated by the resource configuration strategy is higher than the current system resource configuration, the operating system increases the current system resource configuration to the target system resource configuration indicated by the resource configuration strategy, Thereby improving the running quality of the game application.

In summary, in the embodiment of the present application, during the running of the game application, by taking screenshots of the game screen and identifying the game screenshots by the pre-trained game scene recognition model, the game scene corresponding to the game screen is determined, thereby According to the resource configuration strategy corresponding to the game scenario, allocate system resources to the game application, thereby optimizing the performance and power consumption of the game application in different game scenarios; in the resource configuration method provided by the embodiments of the present application, the game application does not need to be accessed The terminal operating system, the terminal operating system can also intelligently analyze the current game scene, thereby eliminating the process of establishing a data channel between the game application program and the operating system, and helping to improve the security of the operating system.

Optionally, the game screenshot is identified by the game scene recognition model to determine the game scene corresponding to the game screen, including:

Get the game type corresponding to the game application;

Enter the game type and game screenshot into the game scene recognition model to obtain the game scene output by the game scene recognition model;

Determine the resource allocation strategy corresponding to the game scene, including:

According to the game type and game scene, determine the resource allocation strategy corresponding to the game scene;

Among them, the game types include at least one of the following: first-person shooter game FPS, third-person shooter game TPS, multiplayer online tactical competitive game MOBA, card game, racing game, music game, simulation management game, adventure game AVG, fighting Game FTG, role-playing game RPG and sandbox game.

Optionally, before determining the resource configuration strategy corresponding to the game scenario, the method further includes:

Get the current remaining power and the application size of the game application. The application size is used to indicate the storage space occupied by the game application;

According to the game type and game scenario, determine the resource allocation strategy corresponding to the game scenario, including:

Input the current remaining power, application size, game type, and game scene into the resource configuration model to obtain the resource configuration strategy output by the resource configuration model.

Optionally, the resource configuration model is a neural network model based on deep learning. The resource configuration model is trained based on sample input parameters and sample output parameters. The sample training parameters include sample power, sample application size, sample type and sample game scene The output parameters include sample resource allocation strategies.

Optionally, take a screenshot of the game screen, and before obtaining the screenshot of the game, the method further includes:

Obtain the application identification of the game application;

Detect whether the application ID belongs to the preset game list. The preset game list contains the identification of the target game application. The target game application sends scene information to the operating system through the data channel with the operating system during operation. The operating system uses Determine the resource allocation strategy according to the game scene indicated by the scene information;

If the application identifier does not belong to the preset game list, then perform a step of taking a screenshot of the game screen to obtain a screenshot of the game.

Optionally, the method further includes:

When receiving the scene information sent by the target game application, take a screenshot of the current game screen of the target game application to obtain a sample game screenshot;

According to the game scene indicated by the scene information and the sample game screenshots, the game scene recognition model is trained through the back propagation algorithm.

Optionally, take a screenshot of the game screen to obtain a screenshot of the game, including:

Display acceleration controls on the upper layer of the game screen;

When the trigger signal for the acceleration control is received, a screenshot is taken of the game screen to obtain a screenshot of the game.

Optionally, after allocating system resources to the game application according to the resource configuration strategy, the method further includes:

Identify the current game scene corresponding to the current game screen every predetermined time interval;

If the current game scene changes, the resource configuration strategy corresponding to the current game scene is re-determined.

Optionally, the resource types of the system resources include at least one of central processing unit CPU resources, graphics processor GPU resources, disk resources, memory resources, and network resources.

Due to the different screen characteristics of game screens in different types of game applications, for example, in the game playing scene, the card game game screen contains card elements, and the racing game game screen contains track elements. Therefore, in order to improve the accuracy of game scene recognition, in a possible implementation manner, a sample game screenshot marked with a game type and a game scene identifier is used to train a game scene recognition model; accordingly, a game scene recognition model is used to identify a game scene At this time, the operating system takes the game type and game screenshot of the current game application as model input, and inputs it into the game scene recognition model. The following uses an exemplary embodiment for description.

Please refer to FIG. 8, which shows a flowchart of a resource configuration method shown in another exemplary embodiment of the present application. This embodiment is exemplified by the method applied to the terminal 100 shown in FIG. 1. The method includes:

Step 801, when the game application is running, the game screen is displayed.

For the implementation of this step, reference may be made to the above step 701, and this embodiment will not be repeated here.

Step 802: Obtain the application identification of the game application.

When a target game application that can establish a data channel with the operating system and perform data communication is installed in the terminal, during the running of such target game application, the scene information (including game scenes) can be reported to the operating system through the data channel , The operating system can configure system resources for the target game application based on the obtained scene information. Therefore, for the target game application, the operating system does not need to perform screenshot and game screenshot identification.

In order to avoid taking screenshots and identifying scenes of the target game application, resulting in a waste of terminal processing resources, in a possible implementation manner, the operating system obtains the application identification of the currently running game application. The application identifier may be the application package name of the game application program.

Step 803, it is detected whether the application identifier belongs to the preset game list, and the preset game list contains the identifier of the target game application program, and the target game application program sends scene information to the operating system through a data channel with the operating system during operation, The operating system is used to determine the resource allocation strategy according to the game scene indicated by the scene information.

Optionally, a preset game list is stored in the terminal, and the preset game list includes the identification of the target game application. Among them, the preset game list is the terminal regularly updates and downloads from the background server.

The terminal operating system detects whether the application ID of the current game application belongs to the preset game list. If it belongs, it determines that the current game application is the target application, and subsequent screenshot scene recognition is not required; if it does not, it determines that the current game application cannot Send the scene information directly to the operating system, and perform step 804.

In other possible implementation manners, when the application identifier does not belong to the preset game list, the operating system further detects whether the application identifier belongs to the game whitelist, and if it belongs to the game whitelist, step 804 is executed. Among them, the game whitelist contains game applications optimized for game performance.

Step 804, if the application identifier does not belong to the preset game list, then take a screenshot of the game screen to obtain a screenshot of the game.

For the step of taking screenshots by the terminal operating system, reference may be made to the above step 702, and this embodiment will not repeat them here.

Step 805: Obtain the game type corresponding to the game application.

In order to improve the accuracy of subsequent game scene recognition, the operating system obtains the game type corresponding to the game application. The game type may include at least one of the following: First-Person Shooter (First-Person Shooter game, FPS), third-person shooter game ( Third-Person Shooter game (TPS), multiplayer online tactical competitive game (Multiplayer Online Battle Arena, MOBA), card game, racing game, music game, simulation management game, adventure game (Adventure Game, AVG), fighting game (Fighting Game, FTG), Role-Playing Game (RPG) and sandbox games.

In a possible implementation manner, the operating system obtains the application name or application package name of the game application program, so as to query the game type of the game application program according to the application name or application package name. The embodiment of the present application does not limit the specific way of acquiring the game type.

Step 806: Enter the game type and game screenshot into the game scene recognition model to obtain the game scene output from the game scene recognition model.

Further, during game scene recognition, the operating system inputs the game type and game screenshot together into the game scene recognition model, thereby obtaining several candidate game scenes and their probabilities output by the game scene recognition model, and the candidate game corresponding to the highest probability The scene is determined to be the game scene corresponding to the current game screen.

Step 807, according to the game type and the game scene, determine the resource configuration strategy corresponding to the game scene.

In the same game scenario, due to different game types, game applications have different requirements on system resources. For example, in the same game scenario, card games have lower system resource requirements than MOBA games. Therefore, operate The system also determines the resource allocation strategy corresponding to the current game scene based on the game type and game scene.

In a possible implementation manner, the terminal stores a correspondence between the three types of games, game scenes, and resource configuration strategies, and the terminal operating system determines the resource configuration strategy corresponding to the current game scene based on the correspondence. Schematically, the corresponding relationship is shown in Table 2.

Table II

game type	Game scene	Resource allocation strategy
card game	Game loading scene	Resource allocation strategy A1
MOBA games	Game loading scene	Resource allocation strategy A2
card game	Game progress scene	Resource allocation strategy C1
MOBA games	Game progress scene	Resource allocation strategy C2

Illustratively, when the identified game scene is "game progress scene" and the game type is "MOBA game", the terminal operating system determines that the resource configuration strategy is "resource configuration strategy C2" based on Table 2.

Because the level of system resource allocation indicated by the resource allocation strategy is also related to the game size (the larger the game size, the higher the demand for system resources), and is related to the current power of the terminal (when the power is low, the system resource configuration is too high Will affect the endurance of the terminal), so in order to further improve the accuracy and applicability of the determined resource allocation strategy, in another possible implementation manner, before determining the resource allocation strategy corresponding to the game scenario, the terminal operating system obtains the current remaining The amount of power and the application size of the game application. The application size is used to indicate the storage space occupied by the game application. When determining the resource configuration strategy corresponding to the game scenario, the terminal operating system inputs the current remaining power, application size, game type, and game scenario into the resource configuration model to obtain the resource configuration strategy output by the resource configuration model.

Optionally, the resource configuration model is a neural network model based on deep learning, and the sample input parameters used in training the resource configuration model include sample power, sample application size, sample type, and sample game scene, and sample output parameters include samples Resource allocation strategy. The embodiments of the present application do not limit the training process of the resource allocation model.

Schematically, the terminal operating system obtains the current remaining power is "50%", the application size is "1.5GB", the game type is "MOBA game" and the game scene is "game progress scene", and the above parameters are entered into the resource configuration After the model, the resource allocation strategy output by the resource allocation model is obtained.

It should be noted that the terminal may also use other parameters related to the resource configuration strategy as model inputs for model training, so as to improve the accuracy of the resource configuration strategy output by the resource configuration model.

Step 808: Assign system resources to the game application according to the resource configuration strategy.

For the implementation of this step, reference may be made to step 705 described above, and this embodiment will not be repeated here.

In this embodiment, considering that different types of games have different game screen characteristics, by entering the game type of the current game application and the game screenshot together into the game scene recognition model, the game scene output by the game scene recognition model is obtained, which improves the game scene recognition Accuracy.

At the same time, considering the different needs of different types of games for system resources, the resource configuration strategy corresponding to the game scenario is determined according to the current game application type and the current game scenario, which improves the accuracy of the determined resource configuration strategy.

In order to further improve the recognition accuracy of the game scene recognition model, in a possible implementation manner, the operating system performs reverse training on the game scene recognition model according to the game screen when the target application reports the scene information. On the basis of FIG. 8, as shown in FIG. 9, after step 808, the following steps may also be included.

Step 809: When receiving the scene information sent by the target game application, take a screenshot of the current game screen of the target game application to obtain a sample game screenshot.

In a possible implementation manner, when detecting whether the application identifier belongs to the preset game list, the terminal operating system receives scene information containing game scenes sent by the target game application during operation of the target game application, and based on the scene The information configures the resource of the target game application; at the same time, the operating system takes a screenshot of the current game screen to obtain a sample game screenshot corresponding to the current game scene of the target game application.

Step 810: Train the game scene recognition model through a back propagation algorithm according to the game scene indicated by the scene information and the sample game screenshots.

Further, the terminal operating system performs back propagation training on the original game scene recognition model according to the game scene indicated by the scene information and sample game screenshots, so as to improve the recognition accuracy of the game scene recognition model.

Optionally, the terminal operating system may perform back propagation training on the game scene recognition model according to the game scene indicated by the scene information, the game type of the target game application, and the sample game screenshot.

In a possible implementation manner, the terminal operating system inputs the sample game screenshots into the original game scene recognition model to obtain several candidate game scenes and their probabilities output by the model, and based on the game scene and various candidates indicated by the scene information Probability corresponding to the game scene, back propagation training.

In this embodiment, when receiving the scene information reported by the target game application, the terminal operating system takes a screenshot of the game screen, so that the game scene recognition model is back-propagated according to the game scene indicated by the game screenshot and the scene information, to This improves the recognition accuracy of the game scene recognition model.

In a possible implementation manner, based on FIG. 8, as shown in FIG. 10, step 804 may include the following steps:

Step 804A: Display the acceleration control.

In a possible implementation manner, the terminal operating system displays the acceleration control and receives the trigger operation on the acceleration control. The acceleration control may be a floating icon located on the upper layer of the game screen, or a shortcut function icon displayed in the shortcut bar (such as an acceleration icon displayed in the opposite area of the display screen), and the trigger operation may be a click operation, a press operation, or a long Press at least one of the operations. This embodiment of the present application does not limit this.

Illustratively, as shown in FIG. 11, an acceleration control 1102 is displayed on the upper layer of the game screen 1101.

Step 804B: When receiving the trigger signal for the acceleration control, take a screenshot of the game screen to obtain a screenshot of the game.

When receiving the trigger signal for the acceleration control, the terminal determines that it is necessary to optimize the performance of the game application, so as to perform operations such as subsequent screenshots and game scene recognition. For a specific implementation manner, reference may be made to the embodiment in FIG. 8, and this embodiment will not be repeated here.

Since the game scene is constantly changing during the game running, for example, after the game is over, the game scene will be switched to the game main interface scene, so the operating system needs to debug the resource configuration strategy in time, accordingly, as shown in FIG. 10, step 808 It also includes the following steps.

Step 811: Identify the current game scene corresponding to the current game picture every predetermined time interval.

In a possible implementation manner, the terminal operating system re-identifies the current game scene corresponding to the current game screen through a screenshot and model recognition operation at predetermined intervals, and detects whether the current game scene and the last recognized game scene If they are consistent, the current resource configuration mode is maintained; if they are not consistent, the resource configuration strategy corresponding to the current game scene is re-determined through the following step 812.

Optionally, the predetermined time interval may be determined according to the predicted duration corresponding to the last recognized game scene. For example, for a MOBA game, when the last game scene identified is a "game progress scene", the terminal operating system determines that the predetermined time interval is 30 minutes.

Step 812, if the current game scene changes, then re-determine the resource configuration strategy corresponding to the current game scene.

When the game scene changes, the terminal operating system re-determines the resource configuration strategy and performs system resource configuration according to the re-determined resource configuration strategy.

In this embodiment, the terminal operating system automatically recognizes the current game screen every predetermined time interval. When the game scene changes, the resource allocation strategy corresponding to the current game scene is re-determined, which can automatically optimize the performance of the game application and reduce The number of user operations to optimize the user experience.

Please refer to FIG. 12, which shows a structural block diagram of a resource configuration apparatus provided by an embodiment of the present application. The resource configuration device can be implemented as all or part of the terminal through software, hardware, or a combination of both. The device includes:

The display module 1201 is used to display a game screen when running a game application;

The screenshot module 1202 is used to take a screenshot of the game screen to obtain a screenshot of the game;

The scene recognition module 1203 is used to recognize the game screenshot through the game scene recognition model and determine the game scene corresponding to the game screen, where the game scene is used to indicate the running stage of the game application;

The strategy determination module 1204 is configured to determine a resource configuration strategy corresponding to the game scene;

The resource allocation module 1205 is configured to allocate system resources to the game application according to the resource configuration strategy.

Optionally, the scene recognition module 1203 includes:

A type obtaining unit, configured to obtain a game type corresponding to the game application;

A scene prediction unit, configured to input the game type and the game screenshot into the game scene recognition model to obtain the game scene output by the game scene recognition model;

The strategy determination module 1204 is used to:

Determine the resource configuration strategy corresponding to the game scenario according to the game type and the game scenario;

Among them, the game types include at least one of the following: FPS, TPS, MOBA, card games, racing games, music games, simulation business games, AVG, FTG, RPG, and sandbox games.

Optionally, the device further includes:

A first obtaining module, configured to obtain the current remaining power and the application size of the game application, and the application size is used to indicate the storage space occupied by the game application;

The strategy determination module 1204 is also used to:

Input the current remaining power, the size of the application program, the game type, and the game scene into a resource configuration model to obtain the resource configuration strategy output by the resource configuration model.

Optionally, the resource configuration model is a neural network model based on deep learning. The resource configuration model is trained based on sample input parameters and sample output parameters. The sample training parameters include sample power, sample application size, and sample type. And a sample game scenario, the sample output parameters include a sample resource configuration strategy.

Optionally, the device further includes:

A second obtaining module, configured to obtain the application identification of the game application program;

The detection module is configured to detect whether the application identifier belongs to a preset game list, and the preset game list includes an identifier of a target game application, and the target game application passes data between the operating system during operation The channel sends scene information to the operating system, and the operating system is used to determine the resource configuration strategy according to the game scene indicated by the scene information;

The screenshot module 1202 is configured to execute the step of taking a screenshot of the game screen to obtain a screenshot of the game if the application identifier does not belong to the preset game list.

Optionally, the device further includes:

A sample screenshot module, configured to take a screenshot of the current game screen of the target game application when receiving the scene information sent by the target game application, to obtain a sample game screenshot;

The training module is configured to train the game scene recognition model through a back propagation algorithm according to the game scene indicated by the scene information and the sample game screenshot.

Optionally, the screenshot module 1202 includes:

A control display unit for displaying an acceleration control on the upper layer of the game screen;

The screenshot unit is configured to take a screenshot of the game screen when the trigger signal for the acceleration control is received to obtain the screenshot of the game.

Optionally, the device further includes:

The timing recognition module is used to identify the current game scene corresponding to the current game screen at predetermined time intervals;

The strategy resetting module is used to re-determine the resource configuration strategy corresponding to the current game scene if the current game scene changes.

Optionally, the resource types of the system resources include at least one of CPU resources, GPU resources, disk resources, memory resources, and network resources.

In summary, in the embodiment of the present application, during the running of the game application, by taking screenshots of the game screen and identifying the game screenshots by the pre-trained game scene recognition model, the game scene corresponding to the game screen is determined, thereby According to the resource configuration strategy corresponding to the game scenario, allocate system resources to the game application, thereby optimizing the performance and power consumption of the game application in different game scenarios; in the resource configuration method provided by the embodiments of the present application, the game application does not need to be accessed The terminal operating system, the terminal operating system can also intelligently analyze the current game scene, thereby eliminating the process of establishing a data channel between the game application program and the operating system, and helping to improve the security of the operating system.

In this embodiment, considering that different types of games have different game screen characteristics, by entering the game type of the current game application and the game screenshot together into the game scene recognition model, the game scene output by the game scene recognition model is obtained, which improves the game scene recognition Accuracy.

At the same time, considering the different needs of different types of games for system resources, the resource configuration strategy corresponding to the game scenario is determined according to the current game application type and the current game scenario, which improves the accuracy of the determined resource configuration strategy.

In this embodiment, when receiving the scene information reported by the target game application, the terminal operating system takes a screenshot of the game screen, so that the game scene recognition model is back-propagated according to the game scene indicated by the game screenshot and the scene information, to This improves the recognition accuracy of the game scene recognition model.

An embodiment of the present application further provides a computer-readable medium that stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the resource configuration method described in the foregoing embodiments .

An embodiment of the present application further provides a computer program product that stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the resource configuration method described in the foregoing embodiments.

Those skilled in the art should be aware that in one or more of the above examples, the functions described in the embodiments of the present application may be implemented by hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media includes computer storage media and communication media, where communication media includes any medium that facilitates transfer of a computer program from one place to another. The storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

The above are only preferred embodiments of this application and are not intended to limit this application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of this application should be included in the protection of this application Within range.

Claims (20)

1. A resource configuration method, the method includes:

   When running the game application, display the game screen;

   Take a screenshot of the game screen to obtain a screenshot of the game;

   Identifying the game screenshot through the game scene recognition model to determine the game scene corresponding to the game screen, where the game scene is used to indicate the running stage of the game application;

   Determine a resource allocation strategy corresponding to the game scene;

   Allocate system resources to the game application according to the resource configuration strategy.
2. The method according to claim 1, wherein the identifying the game screenshot through the game scene recognition model to determine the game scene corresponding to the game screen includes:

   Obtain the game type corresponding to the game application;

   Input the game type and the game screenshot into the game scene recognition model to obtain the game scene output by the game scene recognition model;

   The determining the resource configuration strategy corresponding to the game scene includes:

   Determine the resource configuration strategy corresponding to the game scenario according to the game type and the game scenario;

   Among them, the game types include at least one of the following: first-person shooter game FPS, third-person shooter game TPS, multiplayer online tactical competitive game MOBA, card game, racing game, music game, simulation management game, adventure game AVG , Fighting game FTG, role-playing game RPG and sandbox game.
3. The method according to claim 2, wherein before determining the resource configuration strategy corresponding to the game scene, the method further comprises:

   Acquiring the current remaining power and the application size of the game application, where the application size is used to indicate the storage space occupied by the game application;

   The determining the resource configuration strategy corresponding to the game scenario according to the game type and the game scenario includes:

   Input the current remaining power, the size of the application program, the game type, and the game scene into a resource configuration model to obtain the resource configuration strategy output by the resource configuration model.
4. The method according to claim 3, wherein the resource configuration model is a neural network model based on deep learning, and the resource configuration model is trained based on sample input parameters and sample output parameters, and the sample training parameters include sample power, Sample application size, sample type, and sample game scenario, the sample output parameters include sample resource configuration strategies.
5. The method according to any one of claims 1 to 4, wherein, before taking a screenshot of the game screen and before obtaining a screenshot of the game, the method further comprises:

   Obtaining the application identification of the game application;

   Detecting whether the application identifier belongs to a preset game list, and the preset game list includes an identifier of a target game application program, and the target game application program runs to the operation through a data channel with an operating system during operation The system sends scene information, and the operating system is used to determine the resource configuration strategy according to the game scene indicated by the scene information;

   If the application identifier does not belong to the preset game list, the step of taking a screenshot of the game screen to obtain a screenshot of the game is executed.
6. The method of claim 5, wherein the method further comprises:

   When receiving the scene information sent by the target game application, take a screenshot of the current game screen of the target game application to obtain a sample game screenshot;

   According to the game scene indicated by the scene information and the sample game screenshot, the game scene recognition model is trained through a back propagation algorithm.
7. The method according to claim 5, wherein the screenshot of the game screen to obtain a screenshot of the game includes:

   Display acceleration controls on the upper layer of the game screen;

   When receiving the trigger signal for the acceleration control, take a screenshot of the game screen to obtain the screenshot of the game.
8. The method according to claim 5, wherein after allocating system resources to the game application according to the resource configuration strategy, the method further comprises:

   Identify the current game scene corresponding to the current game screen every predetermined time interval;

   If the current game scene changes, the resource configuration strategy corresponding to the current game scene is re-determined.
9. The method according to any one of claims 1 to 4, wherein

   The resource types of the system resources include at least one of central processing unit CPU resources, graphics processor GPU resources, disk resources, memory resources, and network resources.
10. A resource configuration device, the device includes:

    The display module is used to display the game screen when running the game application;

    A screenshot module, used to take a screenshot of the game screen to obtain a screenshot of the game;

    A scene recognition module, used to identify the game screenshot through the game scene recognition model, and determine the game scene corresponding to the game screen, where the game scene is used to indicate the running stage of the game application;

    A strategy determination module for determining a resource configuration strategy corresponding to the game scene;

    The resource allocation module is configured to allocate system resources to the game application according to the resource configuration strategy.
11. The apparatus according to claim 10, wherein the scene recognition module comprises:

    A type obtaining unit, configured to obtain a game type corresponding to the game application;

    A scene prediction unit, configured to input the game type and the game screenshot into the game scene recognition model to obtain the game scene output by the game scene recognition model;

    The strategy determination module is used to:

    Determine the resource configuration strategy corresponding to the game scenario according to the game type and the game scenario;

    Among them, the game types include at least one of the following: FPS, TPS, MOBA, card games, racing games, music games, simulation business games, AVG, FTG, RPG, and sandbox games.
12. The device according to claim 11, wherein the device further comprises:

    A first obtaining module, configured to obtain the current remaining power and the application size of the game application, and the application size is used to indicate the storage space occupied by the game application;

    The strategy determination module is also used to:

    Input the current remaining power, the size of the application program, the game type, and the game scene into a resource configuration model to obtain the resource configuration strategy output by the resource configuration model.
13. The apparatus according to claim 12, wherein the resource configuration model is a neural network model based on deep learning, and the resource configuration model is trained based on sample input parameters and sample output parameters, and the sample training parameters include sample power, Sample application size, sample type, and sample game scenario, the sample output parameters include sample resource configuration strategies.
14. The device according to any one of claims 10 to 13, wherein the device further comprises:

    A second obtaining module, configured to obtain the application identification of the game application program;

    The detection module is configured to detect whether the application identifier belongs to a preset game list, and the preset game list includes an identifier of a target game application, and the target game application passes data between the operating system during operation The channel sends scene information to the operating system, and the operating system is used to determine the resource configuration strategy according to the game scene indicated by the scene information;

    The screenshot module is configured to execute the step of taking a screenshot of the game screen to obtain a screenshot of the game if the application identifier does not belong to the preset game list.
15. The device of claim 14, wherein the device further comprises:

    A sample screenshot module, configured to take a screenshot of the current game screen of the target game application when receiving the scene information sent by the target game application, to obtain a sample game screenshot;

    The training module is configured to train the game scene recognition model through a back propagation algorithm according to the game scene indicated by the scene information and the sample game screenshot.
16. The apparatus according to claim 14, wherein the screenshot module includes:

    A control display unit for displaying an acceleration control on the upper layer of the game screen;

    The screenshot unit is configured to take a screenshot of the game screen when the trigger signal for the acceleration control is received to obtain the screenshot of the game.
17. The device of claim 14, wherein the device further comprises:

    The timing recognition module is used to identify the current game scene corresponding to the current game screen at predetermined time intervals;

    The strategy resetting module is used to re-determine the resource configuration strategy corresponding to the current game scene if the current game scene changes.
18. The apparatus according to any one of claims 10 to 13, wherein the resource types of the system resources include at least one of CPU resources, GPU resources, disk resources, memory resources, and network resources.
19. A terminal includes a processor and a memory; the memory stores at least one instruction, the at least one instruction is used by the processor to implement the resource configuration according to any one of claims 1 to 9. method.
20. A computer-readable storage medium storing at least one instruction, the at least one instruction being used by a processor to implement the resource configuration method according to any one of claims 1 to 9.

Images