Calculation method for processing touch area of mobile item based on Unreal Engine

Abstract

The invention provides a calculation method for processing a mobile item touch area based on UnreaIEngine, and relates to the technical field of intelligent interaction systems and self-adaptive computation.The calculation method comprises a touch area processing system, and when the calculation method for processing the mobile item touch area based on UnreaIEngine is achieved, processing, controlling and adjusting are conducted through the touch area processing system. The touch area processing system comprises a viewport information acquisition module, a UI information processing module, a touch area control module, an adaptive spatial perception optimization module, a spatial-temporal dynamic modeling module, a super-resolution touch area reconstruction module, a multichannel optimization coexistence module and a curved screen mistaken touch intelligent protection module. According to the mobile item touch area processing and calculating method based on Unreal Engine, the problems of accuracy and flexibility of viewport information acquisition and UI (User Interface) information processing are effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent interactive systems and adaptive computing technologies, and in particular to a calculation method for processing a touch area of ​​a mobile project based on UnrealEngine. Background Art

[0002] The calculation method for touch area processing of mobile projects based on UnrealEngine is mainly to perform unified controllable area click processing for the UI page or viewport control of the mobile project. When the product is running, the interactively generated UI object is obtained and the layout size of the UI in the viewport is calculated, and the UI area or blank area is divided. The touch function can be enabled or disabled according to needs; it can also shield the impact of click operations on other areas that do not need to be clicked or are easy to touch by mistake, so as to avoid user experience problems caused by misoperation. With the rapid development of mobile technology, mobile devices play an increasingly important role in people's daily life and work. As a powerful engine that is widely used in game development and other interactive mobile projects, UnrealEngine provides developers with rich tools and frameworks to build high-quality mobile application experiences.

[0003] In the field of interactive design of mobile projects, touch area processing is one of the core links. Traditional mobile project touch area processing technology has played a certain role in the relatively simple hardware conditions and limited application scenarios of early mobile devices. Generally speaking, its viewport information acquisition mainly relies on the basic screen parameter acquisition function provided by the operating system, which can roughly determine the screen size information, such as obtaining the width and height pixel values ​​of the screen, thereby providing a basic coordinate range for the layout of UI elements. UI information processing is based on pre-set layout rules, and the designed UI elements are placed at the corresponding position of the viewport according to a fixed ratio or absolute coordinates. In some simple applications, such as simple text reading applications or tool applications with relatively simple functions, this method can meet basic touch interaction needs. Users can trigger corresponding functions by touching the area corresponding to the UI element on the screen, such as clicking a button to turn pages or start a tool function.

[0004] Although the existing calculation method for touch area processing can prevent user misoperation from causing product experience problems, there are still many shortcomings: Problem 1: In terms of viewport information acquisition and UI information processing, traditional methods lack flexibility and accuracy. They have poor compatibility with devices of different resolutions, screen ratios, and special screen forms (such as curved screens and folding screens). When processing complex UI layouts, they cannot dynamically adjust the coordinate conversion logic of UI elements according to user touch behavior, which easily leads to a mismatch between the touch position and the UI functional area. For example, in a multi-resolution adapted game, UI elements on a low-resolution device may overlap or have abnormal spacing due to inaccurate coordinate conversion, affecting the user's operating experience and interaction efficiency. Problem 2: In terms of touch area control, the processing capability of non-rectangular UI areas is weak. Traditional technologies make it difficult to achieve accurate touch judgment and function division in non-rectangular UI areas. When faced with irregular architectural model UI interactions or artistically designed non-rectangular UI element combinations in digital twin applications, it is impossible to accurately identify user touch intentions, resulting in frequent misoperations or inability to trigger functions, which limits the innovation of UI design and the diversity of interaction. Problem three: In terms of spatial perception technology, analysis based only on two-dimensional planes can no longer meet the needs of complex applications. It fails to fully explore the nonlinear relationship of touch data in high-dimensional space, resulting in unreasonable priority allocation of touch areas when processing complex UI layouts. For example, in a multi-level, multi-element overlapping game UI interface, it is impossible to intelligently optimize the touch area based on user habits, so that users need to try multiple times to select the target UI element during operation, which reduces the smoothness and accuracy of operation. The traditional method of handling dynamic changes in the touch area has obvious lags and limitations. The predefined event trigger mechanism cannot respond to frequent scene switching, dynamic generation and disappearance of UI elements in real time and intelligently during game operation. For example, in large-scale multiplayer online games, when scene switching or player skill triggering causes UI changes, the touch area is not adjusted in time, which can easily lead to touch unresponsiveness or wrong response, destroying the immersion and continuity of the game; Problem 4: When there are multiple touch modes such as touch screen and external game controller, simple signal allocation and priority setting cannot effectively coordinate resource competition and operation conflicts between channels. For example, in competitive games, when players use touch screen and game controller at the same time, it is easy for the controller operation to interfere with the response of the screen touch area, or the screen touch causes the controller function to be abnormal, and it is impossible to achieve seamless switching and efficient coexistence of multi-channel operations, which limits the innovation of game play and the improvement of user experience.

[0005] Therefore, a calculation method for touch area processing of mobile projects based on UnrealEngine is needed to solve the above problems. Summary of the invention

[0006] Technical issues solved In view of the deficiencies in the prior art, the present invention provides a calculation method for processing the touch area of ​​a mobile item based on UnrealEngine, which solves the problems in the above background technology.

[0007] Technical Solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for calculating the touch area processing of a mobile project based on UnrealEngine, comprising the following steps: S1. Before performing the calculation of the touch area processing of the mobile item, a touch area processing system is provided, and the calculation method is processed, controlled and adjusted by the touch area processing system during implementation; The touch area processing system includes a viewport information acquisition module, a UI information processing module, a touch area control module, an adaptive space perception optimization module, a spatiotemporal dynamic modeling module, a super-resolution touch area reconstruction module, a multi-channel optimization coexistence module, a curved screen mistouch intelligent protection module, an intelligent touch feedback optimization module and a cross-platform touch adaptation module. The touch area processing system acquires and converts viewport coordinate information through the viewport information acquisition module when touch is enabled; the coordinates of the UI window object are converted through the UI information processing module during operation; the touch area control module realizes vertex area touch event processing and touch control of non-rectangular UI areas; the adaptive space perception optimization module divides the touch behavior and UI layout optimization areas based on specific theories; The adaptive spatial perception optimization module adopts tensor decomposition, hypergraph optimization and large-scale spectral graph theory calculation methods to intelligently perceive and optimize the touch area according to the user's touch behavior and the spatial structure of the UI layout. By analyzing the nonlinear relationship in the high-dimensional space and extracting the hidden structure of the touch data based on tensor decomposition, the division and priority allocation of the touch area are optimized to ensure accurate touch under complex layouts. The spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling, and graph convolutional networks to optimize the dynamic adjustment and multi-tasking capabilities of the touch area by modeling the spatiotemporal game environment and user behavior based on the dynamic changes of the touch area and time series data. The super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial network and multi-scale Fourier transform. While improving the accuracy of the low-resolution touch interface, it also provides accurate touch area definition by performing style transfer and image enhancement on the UI area. The touch area processing system acquires and converts viewport coordinate information through the viewport information acquisition module when touch is enabled. The mobile item touch area processing calculation method includes the following steps: S2. After touch control is enabled, the viewport information acquisition module communicates with the underlying display driver of the system to obtain the original pixel information of the screen, converts it into standard viewport coordinates according to preset rules, and establishes a coordinate cache to facilitate subsequent calls; S3. Check the viewport coordinates regularly and update them in time when the screen changes to ensure accurate and real-time coordinates; The coordinates of the UI window object are converted by the UI information processing module during runtime, and the steps are as follows: S4. When the project is running, the UI information processing module monitors the UI generation system and obtains its original layout information when a new UI window is generated; S5. According to the viewport transformation and the scaling factor, the absolute coordinates are converted into viewport 2D coordinates using a linear transformation algorithm and bound to the UI attributes; The touch area control module implements touch event processing in the vertex area and touch control in the non-rectangular UI area. The specific steps are as follows: S6. Processing vertex area touch, first identify the vertex range, and determine whether the point is within the range during touch input. If it is, prevent accidental touch according to the shielding rules, and capture the dragging situation at the same time; S7. For non-rectangular UI areas, the touch event is overloaded, the touch point coordinates and the area geometry information are obtained for collision detection, and if yes, the parameters are passed to control the touch; The adaptive space perception optimization module uses a specific theory to divide the touch behavior and UI layout optimization area, and its execution steps are as follows: S8, project start-up to collect initial touch behavior and UI layout information; S9. Use tensor decomposition theory to analyze high-dimensional relationships, extract hidden structures of touch data, build association graphs, analyze structural characteristics, and complete fine regional division and priority sorting; The spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling, and graph convolutional networks to optimize the dynamic adjustment and multi-tasking capabilities of the touch area by modeling the spatiotemporal game environment and user behavior based on the dynamic changes of the touch area and time series data. The steps are as follows: S10, during game operation, collect the dynamic changes of the touch area and the sequence data of user operation behaviors to build a spatiotemporal data set; s11. Use variational autoencoders to extract touch data features, combine neuron dynamics modeling to predict user touch actions, and use graph convolutional networks to model spatiotemporal relationships and optimize dynamic adjustment strategies; The super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial network and multi-scale Fourier transform. While improving the accuracy of the low-resolution touch interface, it provides accurate touch area definition by performing style transfer and image enhancement on the UI area. It also includes the following steps: When it is found that the resolution of the touch area is low, the module is started to collect low-resolution image data; The neural style transfer reference template is used to improve the aesthetic recognition of the image, and the generative adversarial network is combined to generate high-resolution candidate images. The discriminator compares and optimizes the generator, and finally multi-scale Fourier transform is used to enhance the details and achieve resolution reconstruction optimization.

[0009] Preferably, the mobile item touch area processing calculation method also includes the following steps: Processing the dynamic changes of the touch area and the time series data through the spatiotemporal dynamic modeling module; Improve the touch area resolution and UI display effect by using the super-resolution touch area reconstruction module; Protecting the curved screen from accidental touches by using the intelligent protection module for accidental touches of the curved screen; Dynamically optimizing touch feedback according to user data through the intelligent touch feedback optimization module; The viewport information acquisition module obtains the current game viewport size when the project is started and touch is enabled, and the vertex coordinates are acquired with the upper left corner of the screen as the origin, the entire screen is converted into a specific coordinate format, and the entire interface coordinate information is stored. The UI information processing module receives the generated UI window object during the project operation, and converts it into the corresponding 2D coordinates on the viewport through the attribute calculation of the UI object.

[0010] Preferably, the touch area control module shields touch events and handles drag situations for the area within the vertex range through vertex calculation, and can perform conventional rectangular area control. When processing a non-rectangular UI area, it overloads the touch event of the transferred object, performs parameter judgment on whether to click on the corresponding UI, and passes the judgment parameter to the function to perform touch control of the touch area. The touch area control module includes a function division submodule, which calculates the real-time generated UI for the interactive building objects and interface UI in the exhibition display page of the digital twin direction and the real-time data interaction page of the city and industrial model, and divides the touch area functions according to requirements to achieve unified touch allocation of the overall interface and avoid complex processing caused by multiple touch logics.

[0011] Preferably, the adaptive spatial perception optimization module adopts tensor decomposition, hypergraph optimization and large-scale spectral graph theory to intelligently perceive and optimize the touch area according to the user's touch behavior and the spatial structure of the UI layout. By analyzing the nonlinear relationship in the high-dimensional space and extracting the hidden structure of the touch data based on tensor decomposition, the division and priority allocation of the touch area are optimized to ensure precise touch under complex layouts.

[0012] Preferably, the spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling and graph convolutional networks, targets the dynamic changes of the touch area and time series data, and optimizes the dynamic adjustment and multi-tasking capabilities of the touch area through spatiotemporal modeling of the game environment and user behavior.

[0013] Preferably, the super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial network and multi-scale Fourier transform. While improving the accuracy of the low-resolution touch interface, it provides accurate touch area definition by performing style transfer and image enhancement on the UI area. It is particularly suitable for adaptation to dynamic resolution changes or devices with different resolutions.

[0014] Preferably, the multi-channel optimized coexistence module establishes an intelligent game relationship between multiple operating systems and UI logics by combining dynamic time warping, topology-based optimization algorithms and game theory, and optimizes area allocation and task coexistence in multiple touch modes. The multi-channel optimized coexistence module realizes efficient coexistence and resource allocation between multi-channel operations by adaptively adjusting task weights and touch priorities. The multi-channel optimized coexistence module contains a coexistence logic processing submodule. The coexistence logic processing submodule aims at the coexistence of UI and game movement and viewport operation logic in the game that does not cover the full-screen interface, and uniformly allocates touchable areas and shielded areas through calculation, without locking logic control of the character camera and overloading most touch events in the UI, so as to realize specific operation logic coexistence.

[0015] Preferably, the intelligent protection module for accidental touches of the curved screen combines distributed reinforcement learning, low-rank matrix reconstruction and multimodal perception algorithms, dynamically adjusts the accidental touch area by sensing the curved screen of the mobile phone device and the user's touch habits, and intelligently protects the curved screen area. The intelligent protection module for accidental touches of the curved screen adjusts the touch area through distributed reinforcement learning, so that the touch response is more adapted to the user's personal behavior pattern.

[0016] Preferably, the intelligent touch feedback optimization module uses deep reinforcement learning, adaptive convolutional neural network and dynamic fuzzy control algorithm to dynamically adjust touch feedback parameters according to the user's real-time touch operation and historical behavior data to improve the accuracy and fluency of touch feedback and adapt to each user's touch habits and scenario requirements.

[0017] Preferably, the cross-platform touch adaptation module uses deep learning adaptive algorithms, generative adversarial networks and swarm intelligence optimization algorithms to automatically adapt the touch area settings and interaction logic for different types of touch devices such as smartphones, tablets and game controllers, ensuring a consistent and optimized touch experience on all device platforms.

[0018] Beneficial Effects

[0019] The present invention provides a calculation method for processing the touch area of ​​a mobile project based on UnrealEngine, which has the following beneficial effects: 1. The present invention is based on the calculation method for processing touch area of ​​mobile projects in UnrealEngine, which effectively solves the accuracy and flexibility problems of viewport information acquisition and UI information processing. Through the advanced viewport information acquisition module, it can accurately identify the characteristics of devices with different resolutions, screen ratios and special screen forms, and dynamically adjust the coordinate conversion logic of UI elements according to user touch behavior under complex UI layout. This allows UI elements to be accurately positioned on conventional screens, curved screens, and folding screen devices, avoiding problems such as overlap or abnormal spacing caused by inaccurate coordinate conversion. In multi-resolution adapted games, players can obtain the same UI layout rationality as high-resolution devices on low-resolution devices, greatly improving the user's operating experience and interaction efficiency, ensuring that the user's touch operation can accurately correspond to the desired UI function area, reducing misoperation, and making the interaction process smoother and more natural.

[0020] 2. In terms of touch area control, the calculation method of the present invention shows excellent performance in processing non-rectangular UI areas. For irregular building model UI interactions or non-rectangular UI element combinations with artistic designs in digital twin applications, the function division submodule can deeply analyze their structures and functional requirements to achieve accurate touch judgment and function division. When users operate such non-rectangular UI areas, the system can accurately identify touch intentions, effectively avoid misoperations and ensure normal function triggering. This not only improves the user's operation accuracy in complex UI design scenarios, but also provides UI designers with a broader creative space, encouraging them to break through the limitations of traditional rectangular UIs and boldly carry out innovative designs, thereby promoting the diversified development of mobile project UI design and enriching users' visual and interactive experience.

[0021] 3. The present invention has made significant progress in the field of spatial perception technology. By adopting advanced methods such as tensor decomposition, hypergraph optimization and large-scale spectral graph theory, the nonlinear relationship of touch data in high-dimensional space is deeply explored, and the touch area priority allocation under complex UI layout can be intelligently optimized according to user habits. In a multi-level, multi-element overlapping game UI interface, users can accurately select the target UI element without multiple attempts, which significantly improves the operation fluency and accuracy. At the same time, the spatiotemporal dynamic modeling module effectively overcomes the lag and limitations of traditional touch area dynamic change processing. In large-scale multiplayer online games, in the face of UI changes caused by scene switching or player skill triggering, the touch area can be adjusted in real time and intelligently to ensure timely and accurate touch response, seamlessly connect the game process, greatly enhance the immersion and coherence of the game, and let players immerse themselves in the game world.

[0022] 4. The calculation method in the present invention provides a complete solution. The multi-channel optimization coexistence module abandons the simple signal allocation and priority setting methods. By combining dynamic time warping, topology-based optimization algorithms and game theory, an intelligent game relationship is established between the touch screen and multiple channels such as an external game controller, and task weights and touch priorities are adaptively adjusted. In competitive game scenarios, players can freely combine touch screen and game controller operations to achieve seamless switching and efficient coexistence of multi-channel operations. For example, players can accurately control the character's perspective through the touch screen, and use the game controller to smoothly perform character skill actions. The two do not interfere with each other and work together to enhance efficiency. This not only avoids resource competition and operation conflicts, but also opens up new possibilities for game play innovation, inspiring game developers to design more creative and challenging game modes, thereby comprehensively improving users' gaming experience and fun in multi-channel interaction scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a specific flow chart of the present invention; Figure 2 It is a system operation step diagram of the present invention; Figure 3 A touch area layout diagram of a game viewport of the present invention; Figure 4 It is an overall simulation diagram of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Specific embodiment one:

[0026] like Figure 1-4 As shown, a calculation method for touch area processing of a mobile project based on UnreaIEngine includes a touch area processing system. The calculation method for touch area processing of a mobile project based on UnreaIEngine is processed, controlled and adjusted by the touch area processing system during implementation. The touch area processing system includes a viewport information acquisition module, a UI information processing module, a touch area control module, an adaptive space perception optimization module, a spatiotemporal dynamic modeling module, a super-resolution touch area reconstruction module, a multi-channel optimization coexistence module, a curved screen mistouch intelligent protection module, an intelligent touch feedback optimization module and a cross-platform touch adaptation module. The modules in the touch area processing system work together to realize the function of the calculation method. The viewport information The acquisition module is responsible for acquiring and converting the viewport coordinate information when touch is enabled; the UI information processing module focuses on the coordinate conversion of UI window objects at runtime; the touch area control module implements vertex area touch event processing and non-rectangular UI area touch control; the adaptive space perception optimization module uses specific theories to optimize area division based on touch behavior and UI layout; the spatiotemporal dynamic modeling module uses relevant technologies to process dynamic changes in touch areas and time series data; the super-resolution touch area reconstruction module improves the touch area resolution and UI display effect; the curved screen accidental touch intelligent protection module uses relevant technologies to protect the curved screen from accidental touches; the intelligent touch feedback optimization module dynamically optimizes touch feedback based on user data; the cross-platform touch adaptation module uses algorithms to adapt touch devices.

[0027] The viewport information acquisition module obtains the current game viewport size when the project is started and touch is enabled, and obtains the vertex coordinates with the upper left corner of the screen as the origin, converts the entire screen into a specific coordinate form, and stores the coordinate information of the entire interface. The UI information processing module receives the generated UI window object during the project operation, and converts it into the corresponding 2D coordinates on the viewport through the attribute calculation of the UI object. The touch area control module shields the touch event and handles the drag situation of the area within the vertex range through vertex calculation, and can perform conventional rectangular area control. When processing non-rectangular UI areas, it overloads the touch event of the passing object, performs parameter judgment on whether to click on the corresponding UI, and passes the judgment parameter to the function for touch control of the touch area. The touch area control module contains a functional division submodule. The functional division submodule calculates the real-time generated UI for the interactive building objects and interface UI in the exhibition display page of the digital twin direction and the real-time data interaction page of the city and industrial model, and divides the touch area function according to the requirements to achieve unified touch distribution of the overall interface and avoid complex processing caused by multiple touch logics.

[0028] The adaptive spatial perception optimization module adopts tensor decomposition, hypergraph optimization and large-scale spectrum graph theory to intelligently perceive and optimize the touch area according to the spatial structure of user touch behavior and UI layout. By analyzing the nonlinear relationship in high-dimensional space and extracting the hidden structure of touch data based on tensor decomposition, it optimizes the division and priority allocation of touch areas to ensure accurate touch under complex layouts. The spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling and graph convolutional networks. According to the dynamic changes of touch areas and time series data, it optimizes the dynamic adjustment and multi-tasking processing capabilities of touch areas through spatiotemporal modeling of game environments and user behaviors. The super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial networks and multi-scale Fourier transforms. While improving the accuracy of low-resolution touch interfaces, it provides accurate touch area definition by performing style transfer and image enhancement on UI areas. It is especially suitable for dynamic resolution changes or adaptation of devices with different resolutions.

[0029] The multi-channel optimized coexistence module establishes an intelligent game relationship between multiple operating systems and UI logic by combining dynamic time warping, topology-based optimization algorithms and game theory, and optimizes area allocation and task coexistence in multiple touch modes. The multi-channel optimized coexistence module realizes efficient coexistence and resource allocation between multi-channel operations by adaptively adjusting task weights and touch priorities. The multi-channel optimized coexistence module contains a coexistence logic processing submodule. The coexistence logic processing submodule aims at the coexistence of UI and game movement and viewport operation logic in the game that does not cover the full-screen interface. By capturing the touchable area of ​​the entire interface and uniformly allocating the touchable area and the shielded area after calculation, there is no need to lock the logic control of the character camera and overload most touch events in the UI, so as to realize the coexistence of specific operation logic.

[0030] The intelligent protection module for accidental touches on the curved screen combines distributed reinforcement learning, low-rank matrix reconstruction and multimodal perception algorithms. By sensing the curved screen of the mobile device and the user's touch habits, it dynamically adjusts the accidental touch area and intelligently protects the curved screen area. The intelligent protection module for accidental touches on the curved screen adjusts the touch area through distributed reinforcement learning, so that the touch response is more adapted to the user's personal behavior pattern. The intelligent touch feedback optimization module uses deep reinforcement learning, adaptive convolutional neural networks and dynamic fuzzy control algorithms to dynamically adjust touch feedback parameters based on the user's real-time touch operations and historical behavior data to improve the accuracy and smoothness of touch feedback and adapt to each user's touch habits and scenario requirements. The cross-platform touch adaptation module uses deep learning adaptive algorithms, generative adversarial networks and swarm intelligence optimization algorithms to automatically adapt the touch area settings and interaction logic for different types of touch devices such as smartphones, tablets and game controllers, ensuring a consistent and optimized touch experience on all device platforms.

[0031] When touch is enabled, the touch area processing system obtains and converts the viewport coordinate information through the viewport information acquisition module. The specific steps are as follows: S1. Before performing the calculation of the touch area processing of the mobile item, a touch area processing system is provided, and the calculation method is processed, controlled and adjusted by the touch area processing system during implementation; The touch area processing system includes a viewport information acquisition module, a UI information processing module, a touch area control module, an adaptive space perception optimization module, a spatiotemporal dynamic modeling module, a super-resolution touch area reconstruction module, a multi-channel optimization coexistence module, a curved screen mistouch intelligent protection module, an intelligent touch feedback optimization module and a cross-platform touch adaptation module. The touch area processing system acquires and converts viewport coordinate information through the viewport information acquisition module when touch is enabled; the coordinates of the UI window object are converted through the UI information processing module during operation; the touch area control module realizes vertex area touch event processing and touch control of non-rectangular UI areas; the adaptive space perception optimization module divides the touch behavior and UI layout optimization areas based on specific theories; The adaptive spatial perception optimization module adopts tensor decomposition, hypergraph optimization and large-scale spectral graph theory calculation methods to intelligently perceive and optimize the touch area according to the user's touch behavior and the spatial structure of the UI layout. By analyzing the nonlinear relationship in the high-dimensional space and extracting the hidden structure of the touch data based on tensor decomposition, the division and priority allocation of the touch area are optimized to ensure accurate touch under complex layouts. The spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling, and graph convolutional networks to optimize the dynamic adjustment and multi-tasking capabilities of the touch area by modeling the spatiotemporal game environment and user behavior based on the dynamic changes of the touch area and time series data. The super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial network and multi-scale Fourier transform. While improving the accuracy of the low-resolution touch interface, it also provides accurate touch area definition by performing style transfer and image enhancement on the UI area. The touch area processing system acquires and converts viewport coordinate information through the viewport information acquisition module when touch is enabled. The mobile item touch area processing calculation method includes the following steps: S2. After touch control is enabled, the viewport information acquisition module communicates with the underlying display driver of the system to obtain the original pixel information of the screen, converts it into standard viewport coordinates according to preset rules, and establishes a coordinate cache to facilitate subsequent calls; S3. Check the viewport coordinates regularly and update them in time when the screen changes to ensure accurate and real-time coordinates; The coordinates of the UI window object are converted by the UI information processing module during runtime, and the steps are as follows: S4. When the project is running, the UI information processing module monitors the UI generation system and obtains its original layout information when a new UI window is generated; S5. According to the viewport transformation and the scaling factor, the absolute coordinates are converted into viewport 2D coordinates using a linear transformation algorithm and bound to the UI attributes; The touch area control module implements touch event processing in the vertex area and touch control in the non-rectangular UI area. The specific steps are as follows: S6. Processing vertex area touch, first identify the vertex range, and determine whether the point is within the range during touch input. If it is, prevent accidental touch according to the shielding rules, and capture the dragging situation at the same time; S7. For non-rectangular UI areas, the touch event is overloaded, the touch point coordinates and the area geometry information are obtained for collision detection, and if yes, the parameters are passed to control the touch; The adaptive space perception optimization module uses a specific theory to divide the touch behavior and UI layout optimization area, and its execution steps are as follows: S8, project start-up to collect initial touch behavior and UI layout information; S9. Use tensor decomposition theory to analyze high-dimensional relationships, extract hidden structures of touch data, build association graphs, analyze structural characteristics, and complete fine regional division and priority sorting; The spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling, and graph convolutional networks to optimize the dynamic adjustment and multi-tasking capabilities of the touch area by modeling the spatiotemporal game environment and user behavior based on the dynamic changes of the touch area and time series data. The steps are as follows: S10, during game operation, collect the dynamic changes of the touch area and the sequence data of user operation behaviors to build a spatiotemporal data set; s11. Use variational autoencoders to extract touch data features, combine neuron dynamics modeling to predict user touch actions, and use graph convolutional networks to model spatiotemporal relationships and optimize dynamic adjustment strategies; The super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial network and multi-scale Fourier transform. While improving the accuracy of the low-resolution touch interface, it provides accurate touch area definition by performing style transfer and image enhancement on the UI area. It also includes the following steps: When it is found that the resolution of the touch area is low, the module is started to collect low-resolution image data; The neural style transfer reference template is used to improve the aesthetic recognition of the image, and the generative adversarial network is combined to generate high-resolution candidate images. The discriminator compares and optimizes the generator, and finally multi-scale Fourier transform is used to enhance the details and achieve resolution reconstruction optimization. Specific embodiment 2: like Figure 1-4 As shown, the key algorithm mentioned in Example 1 is analyzed in detail below, including its core mathematical formula and explanation: 1. Adaptive spatial perception optimization module (1) Tensor decomposition formula:

[0033] in A three-dimensional tensor of touch area data, representing spatial coordinates, touch behavior, and temporal characteristics; : low-rank factors, representing the decomposition basis in three dimensions (space, touch behavior, and time); The rank of the decomposition indicates the complexity of the tensor; : Tensor outer product, used to represent the decomposed data structure. By extracting the hidden structure in the touch data through tensor decomposition, the division and priority allocation of the touch area can be optimized.

[0034] (2) Hypergraph optimization formula:

[0035] in The edge set of the hypergraph represents the association between touch areas; : Super edge The weight of represents the strength of regional association; : Touch area and The characteristic value of The set of vertices of the hypergraph represents the touch area; node The degree.

[0036] By optimizing the hypergraph structure, intelligent perception and optimization of the touch area can be achieved.

[0037] (3) Spectral theory formula:

[0038] in : Laplace matrix, representing the structural relationship of the touch area; : Degree matrix, the diagonal elements are the degrees of the nodes; : Weight matrix, representing the weights between touch areas.

[0039] 2. Spatiotemporal dynamic modeling module Algorithms involved: Variational Autoencoder (VAE), Neuron Dynamics Modeling, Graph Convolutional Network (GCN) (1) Variational Autoencoder Formula:

[0040] in : Input touch area data; : Hidden variable, representing the potential characteristics of the touch area; : Approximate the posterior distribution and learn the latent variables; : Conditional probability distribution, used to generate touch data; KL divergence measures the difference between two distributions.

[0041] The dynamic changes of the touch area are modeled through a variational autoencoder and a latent space representation is generated.

[0042] (2) Graph convolutional network formula:

[0043] in The adjacency matrix of the graph, which represents the connections between touch areas; : degree matrix, the degree of the node; : No. The feature matrix of the layer, which represents the current state of the touch area; : No. The weight matrix of the layer; is the activation function.

[0044]

[0045] in The adjacency matrix of the graph, which represents the connections between touch areas; : degree matrix, the degree of the node; : No. The feature matrix of the layer, which represents the current state of the touch area; : No. The weight matrix of the layer; is the activation function.

[0046] GCN uses the structural information of the graph to perform spatiotemporal modeling of the touch area and process dynamically changing data.

[0047] 3. Super-resolution touch area reconstruction module Algorithms involved: Neural style transfer, generative adversarial network (GAN), multi-scale Fourier transform (1) Generate adversarial network formula:

[0048] in : Generator, used to generate high-resolution images of the touch area; : Discriminator, used to distinguish between real and generated touch areas; : real data; Random noise.

[0049] GAN generates high-resolution touch area images through adversarial training.

[0050] (2) Multi-scale Fourier transform formula:

[0051] in The spatial signal of the touch area; Frequency domain signal; Frequency components; Multi-scale Fourier transform is used to analyze the frequency characteristics of the touch area and enhance image details.

[0052] 4. The curved screen accidentally touches the intelligent protection module Involved algorithms: distributed reinforcement learning, low-rank matrix reconstruction, multimodal perception algorithm (1) Distributed reinforcement learning formula:

[0053] in state Take action the value of Learning rate; Instant rewards; Discount factor; distributed reinforcement learning dynamically adapts to user behavior patterns by adjusting the touch area.

[0054] (2) Low-rank matrix reconstruction formula:

[0055] in Touch area data matrix; Low-rank matrix, representing the normal touch area; Sparse matrix, indicating abnormal mis-touch area; Regularization parameter.

[0056] The false touch area and the normal area are separated by low-rank matrix reconstruction. Specific embodiment three: like Figure 1-4 As shown, the following is a calculation method for touch area processing of a mobile project based on UnrealEngine, with specific application logic steps of each module and algorithm: 1. Initialization and basic information acquisition When the project is started and the touch function is enabled, the viewport information acquisition module starts to work. It accurately obtains the size data of the current game viewport, sets the upper left corner of the screen as the origin (0,0), converts the entire screen range into the corresponding coordinate system according to the established rules, such as (0,0)~(x,x), and properly stores this complete interface coordinate information, building a basic framework for the subsequent precise positioning and processing of the touch area.

[0058] 2. UI information integration and coordinate conversion When the project is running, once a UI window is generated, the UI information processing module will respond immediately. It receives these UI window objects, and deeply analyzes the key attributes of the UI object such as Size or image, and uses a specially designed calculation algorithm to accurately convert it into the corresponding 2D coordinate information in the viewport space. This conversion process enables the UI to have a clear and precise position mapping in the viewport space, providing the necessary data support for subsequent touch operation processing of the UI area.

[0059] 3. Touch area core processing logic The touch area control module works based on the previously acquired viewport coordinate information and the UI's 2D coordinate data. For touch areas of conventional rectangular shapes, it can efficiently handle touch event shielding or drag operation logic through vertex coordinate calculations. For example, in the rectangular button area of ​​the game interface, it can be precisely set to only allow clicks and prohibit dragging, or vice versa, based on demand, thereby achieving fine control over the interactive behavior of specific rectangular touch areas.

[0060] When faced with the touch processing needs of non-rectangular UI areas, the module will overload the processing mechanism by passing the touch events of the object. For non-rectangular interactive building objects or special UI elements in exhibition display pages in digital twin applications and real-time data interaction pages of urban and industrial models, it first performs in-depth calculation and analysis on the real-time generated UI, fully considers its unique spatial structure and functional requirements, and then scientifically and reasonably divides the touch area functions. For example, in the display of architectural models, for the non-rectangular structural parts of the building (such as special-shaped decorative components, irregular functional areas, etc.), precise touchable areas are divided according to their functional characteristics, such as the area for clicking to view detailed information and the drag control area in a specific direction. Then, rigorous parameters are used to determine whether the user clicks on the corresponding UI area, and the judgment result parameters are passed to a special function to achieve precise touch control of the non-rectangular UI area, thereby achieving the unity and efficiency of touch distribution of the entire interface, and effectively avoiding processing difficulties and user misoperations caused by complex and diverse touch logic.

[0061] 4. Adaptive optimization and precise perception The adaptive spatial perception optimization module uses tensor decomposition, hypergraph optimization and large-scale spectral graph theory to intervene in the touch area processing flow. It captures the user's touch behavior data in real time and comprehensively, including the distribution pattern of touch points and the characteristics of touch trajectories. At the same time, it deeply combines the complex spatial structure information of the current UI layout, conducts in-depth analysis of nonlinear relationships in high-dimensional space, and uses tensor decomposition technology to mine the hidden structural information in touch data, such as the user's potential touch intention patterns or preference patterns under a specific UI layout. Based on these in-depth analysis results, the division of touch areas is intelligently optimized, and the touch priority of different areas is reasonably allocated. For example, in a complex game main menu interface, the touch priority of important functional areas is intelligently improved based on the areas that users frequently touch and the habitual operation paths, ensuring that in a complex and diverse UI layout environment, users can perform touch operations with higher accuracy and convenience.

[0062] 5. Spatiotemporal dynamic modeling and response adjustment The spatiotemporal dynamic modeling module integrates variational autoencoders, neuron dynamics modeling, and graph convolutional network technologies to focus on processing the dynamic changes of the touch area over time. For example, during the game, with the dynamic switching of game scenes and the real-time generation or disappearance of UI elements, the touch area will inevitably change accordingly. This module conducts comprehensive spatiotemporal modeling and analysis of the game environment (covering factors that potentially affect touch operations, such as dynamic position changes of objects in the scene and real-time changes in lighting conditions) and user behavior (including continuous touch operation sequences, dynamic changes in operation speed, etc.), accurately predicts possible changes in the touch area, and formulates optimized touch area dynamic adjustment strategies and multi-tasking solutions in advance. For example, in an adventure game with rich scene changes, when the player character enters different areas (such as battle scenes, exploration scenes, and plot interaction scenes), the relevant function buttons and prompt information on the UI will change accordingly. The spatiotemporal dynamic modeling module can timely and accurately adjust the response logic of the touch area to ensure that players can perform touch operations smoothly and unimpeded during various game state transitions, such as quickly switching UI areas corresponding to weapons or skills during battle, and conveniently operating map navigation or item collection UI elements in exploration scenes.

[0063] 6. Super-resolution reconstruction and display enhancement The super-resolution touch area reconstruction module processes the touch area with the help of neural style transfer, generative adversarial network and multi-scale Fourier transform technology. In low-resolution mobile devices or scenarios with special requirements for UI display accuracy, the module deeply processes the image data of the touch area. For example, in the UI interface of a high-definition map navigation application, when the user zooms in and out of the map, the multi-scale Fourier transform is used to carefully analyze and process the touch signals of different frequency components, effectively reconstructing and optimizing the detailed display effect of the touch area. At the same time, the UI area is style transferred and image enhanced through neural style transfer and generative adversarial network technology, so that the key information of roads and places on the map is presented in a clearer and more accurate manner, providing a more precise touch area definition. This processing mechanism ensures that users can get high-quality and consistent touch experience whether in the original fixed resolution scene, during dynamic resolution changes, or running on devices with different resolution specifications.

[0064] 7. Multi-channel coexistence and optimized processing The multi-channel optimized coexistence module combines dynamic time warping, topology-based optimization algorithms and game theory to build an intelligent and coordinated game relationship between multiple operating systems and diverse UI logics. In the context of the diversity of modern mobile devices, there may be multiple touch modes (such as touch screen operation and external game controller operation) and different operating system environments. This module adaptively adjusts the weights and touch priorities of different tasks to achieve efficient and harmonious coexistence and reasonable allocation of resources between multi-channel operations. The coexistence logic processing submodule targets the typical situation in which the UI that does not cover the full-screen interface in the game coexists with the game movement and viewport operation logic. By fully capturing the touchable area information of the entire interface, it conducts in-depth calculation and analysis, and uniformly plans the reasonable allocation of touchable areas and shielded areas. For example, in a large-scale 3D game, there is a small map UI that does not cover the entire screen. While controlling the movement of the game character (game movement operation logic) and rotating the perspective (viewport operation logic), the player can also zoom, mark, and perform other operations on the small map UI (UI operation logic). The multi-channel optimization coexistence module can ensure that these different types of operations do not interfere with each other. There is no need for complex locking logic control of the character camera and tedious overloading of most touch events in the UI, thereby achieving smooth and natural coexistence of specific operation logic, greatly improving the user's operating experience in multi-channel interaction scenarios.

[0065] 8. Intelligent protection mechanism for accidental touch of curved screen The intelligent protection module for accidental touches on curved screens uses distributed reinforcement learning, low-rank matrix reconstruction, and multimodal perception algorithms to effectively deal with the problem of accidental touches that are prone to occur on curved screens of mobile devices in specific scenarios of horizontal screen games. It uses a multimodal perception algorithm to fully perceive the unique physical characteristics of the curved screens of mobile devices and the personalized touch habits of users, such as the user's habitual posture of holding the phone, changes in the contact area of ​​the finger on the curved screen, pressure distribution characteristics, and other multi-dimensional information. It then uses low-rank matrix reconstruction technology to deeply process the touch data and accurately analyze the possible characteristics of accidental touch signals. It then uses a distributed reinforcement learning algorithm to continuously and dynamically adjust the response strategy of the touch area and intelligently adjust the accidental touch area. For example, in the process of horizontal screen games, it can intelligently identify and shield the area on the edge of the curved surface of the mobile phone that is prone to accidental touches due to the user's grip, or flexibly adjust the touch sensitivity of the area according to the user's actual operation, so that the touch response can be highly consistent with the user's personal behavior pattern, effectively reducing the abnormal game experience caused by accidental touches, and improving the stability and fluency of users' game operations on curved screen devices.

[0066] 9. Optimization and improvement of intelligent touch feedback The intelligent touch feedback optimization module uses deep reinforcement learning, adaptive convolutional neural networks and dynamic fuzzy control algorithms to work based on users' real-time touch operation data and historical behavior data. It collects and analyzes users' touch operation details in real time, such as the size of click force, sliding speed and operation frequency, and combines the rich historical behavior data accumulated by users in similar scenarios in the past, such as the habitual operation methods and preference settings in the same type of game scenarios or application operations in the past. It continuously optimizes the touch feedback strategy through deep reinforcement learning algorithms and uses adaptive convolutional neural networks to accurately and dynamically adjust the key parameters of touch feedback, such as flexibly adjusting the intensity of touch vibration, the volume and tone of sound feedback and other multi-dimensional parameters. The dynamic fuzzy control algorithm is then used to dynamically determine the range of parameter adjustment according to different touch scenarios and user personalized preferences, so as to achieve a significant improvement in the accuracy and fluency of touch feedback, highly adapt to each user's unique touch habits and various complex and changeable scenario requirements, thereby comprehensively enhancing the user's experience and immersion in the entire mobile project interaction process.

[0067] 10. Cross-platform adaptation and consistency assurance The cross-platform touch adaptation module uses deep learning adaptive algorithms, generative adversarial networks, and swarm intelligence optimization algorithms to fully adapt to different types of touch devices such as smartphones, tablets, and game controllers. In the process of mobile project development, the module fully considers the significant differences between different devices in screen size, resolution, touch mode (such as pure touch screen operation and physical button operation), etc. When the project is running, the module can automatically and accurately detect the type information of the current device, and then automatically adapt the touch area settings and interaction logic according to the specific characteristics of the device. For example, in a cross-platform game application, the touch operation mode of a smartphone The UI element layout and touch area design are adapted to be suitable for flexible finger clicks and slide operations; for tablet devices, the size ratio and arrangement of UI elements are optimized according to their larger screen display area, and the sensitivity parameters of the touch area are reasonably adjusted; for game controller devices, the various button operation functions of the controller are scientifically and reasonably mapped with the corresponding functions in the game to ensure a consistent and optimized touch experience when running on different device platforms, so that users do not need to re-adapt to complex touch operations when switching to different types of devices to operate mobile projects, which greatly improves the versatility and user-friendliness of mobile projects in a multi-platform environment. Specific embodiment four: like Figure 1-4 As shown, the following is a detailed hardware composition and hardware description of each module in Example 1: 1. Viewport information acquisition module Graphics processing units (GPUs), such as the GPU cores in Qualcomm's Adreno series or Apple's A series chips, provide accurate viewport data with extremely low latency to meet the needs of mobile projects with high real-time requirements; display controllers: are the bridge between the GPU and the display, responsible for managing the transmission of display signals and the setting of display parameters. In the process of obtaining viewport information, the display controller can assist in obtaining information such as the physical size and pixel density of the current display.

[0069] 2. UI information processing module Central Processing Unit (CPU): In UI information processing, the CPU is responsible for parsing the relevant attribute data of the UI window object, such as the Size attribute, and performing coordinate conversion calculations according to a pre-set algorithm.

[0070] Memory (RAM): used to store the attribute data of UI window objects as well as the intermediate data and calculation results in the coordinate conversion process. Sufficient memory capacity and fast read and write speed can ensure the smoothness of UI information processing.

[0071] 3. Touch area control module Touch sensor: There are dense capacitive sensing units distributed under the screen of mobile devices. When the user touches the screen with his finger, it will cause a change in the capacitance value. The touch sensor can accurately detect the position coordinates of the touch point and pass this information to the touch area control module.

[0072] 4. Adaptive spatial perception optimization module High-performance CPU and GPU work together: The CPU is responsible for processing complex logical operations and data management, such as matrix operation control in tensor decomposition algorithms, graph structure construction and update in hypergraph optimization algorithms, etc. The GPU takes advantage of processing large-scale data parallel computing, such as the decomposition calculation of high-dimensional matrices in the process of tensor decomposition, and the feature calculation of graphs in large-scale spectral graph theory analysis.

[0073] 5. Spatiotemporal dynamic modeling module CPU and GPU combination: Similar to the adaptive spatial perception optimization module, the CPU and GPU also undertake different tasks in spatiotemporal dynamic modeling. The CPU is responsible for processing the logical control parts such as model construction and parameter update, such as the construction and training control of the encoder and decoder networks in the variational autoencoder, the setting and updating of model parameters in neuron dynamics modeling, and the definition and management of the graph structure in the graph convolutional network. The GPU plays a role in data processing intensive tasks. For example, in the training process of the variational autoencoder, a large amount of touch area time series data is encoded and decoded, and the feature calculation and information propagation of graph nodes in the graph convolutional network are accelerated.

[0074] 6. Super-resolution touch area reconstruction module Display Processing Unit (DPU): The DPU is responsible for converting the touch area image data after super-resolution reconstruction and style transfer into a format suitable for display screens, and for final image rendering and output. It can perform post-processing such as color correction and contrast adjustment on the image to ensure the best visual effect on the display screen. For example, some DPUs support high dynamic range (HDR) display technology, which can display super-resolution reconstructed images with richer colors and higher contrast, enhancing the user's visual perception of UI elements during touch operations and further optimizing the touch experience.

[0075] 7. Multi-channel optimized coexistence module Multi-channel communication interface: including Bluetooth interface, USB interface and different touch channel signal transmission lines inside the device. These interfaces are responsible for transmitting data between different touch channels to ensure that the touch operation information of each channel can be obtained by the multi-channel optimization coexistence module in a timely and accurate manner.

[0076] 8. Accidental touch of the intelligent protection module on the curved screen Curved screen touch sensor array: The unique touch sensor array of the curved screen not only covers the flat area of ​​the screen, but also extends to the curved part. This special sensor array can accurately detect touch operations on the curved area and provide detailed touch position information for false touch judgment.

[0077] Sensor signal processing chip: This chip is responsible for receiving and processing signals from the curved screen touch sensor array. It performs preliminary filtering and noise reduction on the touch data, extracts valid touch information, and passes it to the subsequent low-rank matrix reconstruction and distributed reinforcement learning modules for false touch judgment and strategy adjustment.

[0078] 9. Intelligent touch feedback optimization module Audio chip and vibration motor: The audio chip is responsible for generating sound signals for touch feedback, such as click sound effects, sliding sound effects, etc. It can adjust the volume, pitch, timbre and other parameters of the sound according to the instructions of the CPU or other control modules to provide rich and diverse audio feedback; the vibration motor provides tactile feedback, and represents different touch events through vibrations of different intensities and frequencies. Specific embodiment five:

[0080] like Figure 4 As shown in the figure, the first heat map shows the weight distribution of the touch area that changes dynamically over time during the optimization process based on the reinforcement learning algorithm and multimodal feature analysis. The change in color of each area reflects the degree to which the algorithm pays attention to different touch areas during the optimization process. For example, the highlighted area represents the area that is optimized first. The heat map combines the multimodal data fusion algorithm and the dynamic multimodal visual analysis algorithm (DMVA), showing how the system extracts the importance of the touch area in the multidimensional feature data, effectively avoiding the over-optimization of a single area, improving the global responsiveness of the touch area, and enhancing the balance of the user experience. This dynamic distribution model is suitable for user interaction devices, such as smartphone touch screens, and adjusts the touch sensitivity and response range by analyzing user behavior data.

[0081] The second line graph shows the performance of spatiotemporal features extracted based on the efficient multi-source data fusion and prediction algorithm (HMD-FPA) during the compression process. The changes in the feature values ​​in the line graph reflect the core information that can still be retained after the different touch areas are compressed, while achieving effective simplification of the data dimension. This process uses the feature selection capability of the dynamic multi-objective irrigation control algorithm (DMO-ICA) to prioritize the extraction of key touch areas, combined with the adaptive adjustment capability of the ITFOA algorithm to achieve compression of data dimensions. The compressed data reduces the complexity of model training and provides an efficient basis for subsequent optimization strategies.

[0082] The third line graph verifies the system's reconstruction capability under multi-objective optimization after compressing the touch area data. The line graph shows that the reconstructed touch area is basically consistent with the original feature trend, indicating that the multimodal data fusion and reconstruction algorithm retains the core information.

[0083] The fourth bar chart shows the Q table trained by the reinforcement learning algorithm, which reflects the allocation priority of the touch area under different strategies. The height of each bar represents the value distribution of different touch areas in the corresponding strategy, reflecting the strategy generation ability of reinforcement learning.

[0084] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprising a reference structure" do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0085] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A calculation method for touch area processing of a mobile project based on UnrealEngine, characterized in that: The following steps are involved: S1. Before performing the calculation of the touch area processing of the mobile item, a touch area processing system is provided, and the calculation method is processed, controlled and adjusted by the touch area processing system during implementation; The touch area processing system includes a viewport information acquisition module, a UI information processing module, a touch area control module, an adaptive space perception optimization module, a spatiotemporal dynamic modeling module, a super-resolution touch area reconstruction module, a multi-channel optimization coexistence module, a curved screen mistouch intelligent protection module, an intelligent touch feedback optimization module and a cross-platform touch adaptation module. The touch area processing system acquires and converts viewport coordinate information through the viewport information acquisition module when touch is enabled; and converts the coordinates of the UI window object through the UI information processing module during operation; The touch area control module implements touch event processing in the vertex area and touch control in the non-rectangular UI area; the adaptive space perception optimization module divides the touch behavior and UI layout optimization area based on specific theories; The adaptive spatial perception optimization module adopts tensor decomposition, hypergraph optimization and large-scale spectral graph theory calculation methods to intelligently perceive and optimize the touch area according to the user's touch behavior and the spatial structure of the UI layout. By analyzing the nonlinear relationship in the high-dimensional space and extracting the hidden structure of the touch data based on tensor decomposition, the division and priority allocation of the touch area are optimized to ensure accurate touch under complex layouts. The spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling, and graph convolutional networks to optimize the dynamic adjustment and multi-tasking capabilities of the touch area by modeling the spatiotemporal game environment and user behavior based on the dynamic changes of the touch area and time series data. The super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial network and multi-scale Fourier transform. While improving the accuracy of the low-resolution touch interface, it provides accurate touch area definition by performing style transfer and image enhancement on the UI area. The touch area processing system acquires and converts viewport coordinate information through the viewport information acquisition module when touch is enabled. The mobile item touch area processing calculation method includes the following steps: S2. After touch control is enabled, the viewport information acquisition module communicates with the underlying display driver of the system to obtain the original pixel information of the screen, converts it into standard viewport coordinates according to preset rules, and establishes a coordinate cache to facilitate subsequent calls; S3. Check the viewport coordinates regularly and update them in time when the screen changes to ensure accurate and real-time coordinates; The coordinates of the UI window object are converted by the UI information processing module during runtime, and the steps are as follows: S4. When the project is running, the UI information processing module monitors the UI generation system and obtains its original layout information when a new UI window is generated; S5. According to the viewport transformation and the scaling factor, the absolute coordinates are converted into viewport 2D coordinates using a linear transformation algorithm and bound to the UI attributes; The touch area control module implements touch event processing in the vertex area and touch control in the non-rectangular UI area. The specific steps are as follows: S6. Processing vertex area touch, first identify the vertex range, and determine whether the point is within the range during touch input. If it is, prevent accidental touch according to the shielding rules, and capture the dragging situation at the same time; S7. For non-rectangular UI areas, the touch event is overloaded, the touch point coordinates and the area geometry information are obtained for collision detection, and if yes, the parameters are passed to control the touch; The adaptive space perception optimization module uses a specific theory to divide the touch behavior and UI layout optimization area, and its execution steps are as follows: S8, project start-up to collect initial touch behavior and UI layout information; S9. Use tensor decomposition theory to analyze high-dimensional relationships, extract hidden structures of touch data, build association graphs, analyze structural characteristics, and complete fine regional division and priority sorting; The spatiotemporal dynamic modeling module combines variational autoencoders, neuron dynamics modeling, and graph convolutional networks to optimize the dynamic adjustment and multi-tasking capabilities of the touch area by modeling the spatiotemporal game environment and user behavior based on the dynamic changes of the touch area and time series data. The steps are as follows: S10, during game operation, collect the dynamic changes of the touch area and the sequence data of user operation behaviors to build a spatiotemporal data set; s11. Use variational autoencoders to extract touch data features, combine neuron dynamics modeling to predict user touch actions, and use graph convolutional networks to model spatiotemporal relationships and optimize dynamic adjustment strategies; The super-resolution touch area reconstruction module reconstructs and optimizes the spatial resolution of the touch area by combining neural style transfer, generative adversarial network and multi-scale Fourier transform. While improving the accuracy of the low-resolution touch interface, it provides accurate touch area definition by performing style transfer and image enhancement on the UI area. It also includes the following steps: When it is found that the resolution of the touch area is low, the module is started to collect low-resolution image data; The neural style transfer reference template is used to improve the aesthetic recognition of the image, and the generative adversarial network is combined to generate high-resolution candidate images. The discriminator compares and optimizes the generator, and finally multi-scale Fourier transform is used to enhance the details and achieve resolution reconstruction optimization.

2. The calculation method for touch area processing of a mobile project based on UnrealEngine according to claim 1, characterized in that: The mobile item touch area processing calculation method also includes the following steps: Processing the dynamic changes of the touch area and the time series data through the spatiotemporal dynamic modeling module; Improve the touch area resolution and UI display effect by using the super-resolution touch area reconstruction module; Protecting the curved screen from accidental touches by using the intelligent protection module for accidental touches of the curved screen; Dynamically optimizing touch feedback according to user data through the intelligent touch feedback optimization module; The viewport information acquisition module obtains the current game viewport size when the project is started and touch is enabled, and the vertex coordinates are acquired with the upper left corner of the screen as the origin, the entire screen is converted into a specific coordinate form, and the entire interface coordinate information is stored. The UI information processing module receives the generated UI window object during the project operation, and converts it into the corresponding 2D coordinates on the viewport through the attribute calculation of the UI object.

3. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The touch area control module shields touch events and handles drag situations in the area within the vertex range through vertex calculation, and can perform conventional rectangular area control. When processing non-rectangular UI areas, it overloads by passing the touch event of the object, performs parameter judgment on whether to click on the corresponding UI, and passes the judgment parameter to the function to perform touch control of the touch area. The touch area control module includes a function division submodule, which calculates the real-time generated UI for the interactive building objects and interface UI in the exhibition display page of the digital twin direction and the real-time data interaction page of the city and industrial model, and divides the touch area functions according to requirements to achieve unified touch distribution of the overall interface and avoid complex processing caused by multiple touch logics.

4. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The algorithm formula of the tensor decomposition is as follows: ， in A three-dimensional tensor of touch area data, representing spatial coordinates, touch behavior, and temporal characteristics; : low-rank factors, representing the decomposition basis in the three dimensions of space, touch behavior and time; The rank of the decomposition indicates the complexity of the tensor; : Tensor outer product, used to represent the decomposed data structure; The formula for the hypergraph optimization is as follows: ； in The edge set of the hypergraph represents the association between touch areas; : Super edge The weight of represents the strength of regional association; : Touch area and The characteristic value of The set of vertices of the hypergraph represents the touch area; node The degree of The spectrum theory formula is as follows: ； in : Laplace matrix, representing the structural relationship of the touch area; : Degree matrix, the diagonal elements are the degrees of the nodes; : Weight matrix, representing the weights between touch areas.

5. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The formula of the variational autoencoder is as follows: ； in : Input touch area data; : Hidden variable, representing the potential characteristics of the touch area; : Approximate the posterior distribution and learn the latent variables; : Conditional probability distribution, used to generate touch data; KL divergence, which measures the difference between two distributions; The graph convolutional network formula is as follows; ； in The adjacency matrix of the graph, which represents the connections between touch areas; : degree matrix, the degree of the node; : No. The feature matrix of the layer, which represents the current state of the touch area; : No. The weight matrix of the layer; is the activation function.

6. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The formula for generating the adversarial network is as follows: ； in : Generator, used to generate high-resolution images of the touch area; : Discriminator, used to distinguish between real and generated touch areas; : real data; Random noise; The multi-scale Fourier transform formula is as follows: ； in : spatial signal of the touch area; Frequency domain signal; Frequency components.

7. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The multi-channel optimization coexistence module combines dynamic time warping, topology-based optimization algorithm and game theory. The multi-channel optimization coexistence module realizes efficient coexistence and resource allocation between multi-channel operations by adaptively adjusting task weights and touch priorities. The multi-channel optimization coexistence module contains a coexistence logic processing submodule. The coexistence logic processing submodule aims at the coexistence of UI, game movement and viewport operation logic that do not cover the full-screen interface in the game. By capturing the touchable area of ​​the entire interface and uniformly allocating the touchable area and the shielded area through calculation, there is no need to lock the logic control of the character camera and overload most touch events in the UI, so as to realize the coexistence of specific operation logic.

8. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The curved screen accidental touch intelligent protection module combines distributed reinforcement learning, low-rank matrix reconstruction and multimodal perception algorithms, dynamically adjusts the accidental touch area by sensing the curved screen of the mobile phone device and the user's touch habits, and intelligently protects the curved screen area. The curved screen accidental touch intelligent protection module adjusts the touch area through distributed reinforcement learning, so that the touch response is more adapted to the user's personal behavior pattern; The distributed reinforcement learning formula is as follows: ； in :state Take action the value of Learning rate; Instant rewards; Discount factor.

9. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The intelligent touch feedback optimization module uses deep reinforcement learning, adaptive convolutional neural network and dynamic fuzzy control algorithm to dynamically adjust touch feedback parameters according to the user's real-time touch operation and historical behavior data to improve the accuracy and fluency of touch feedback and adapt to each user's touch habits and scenario requirements.

10. The calculation method for handling touch area of ​​a mobile project based on UnrealEngine according to claim 1, characterized in that: The cross-platform touch adaptation module uses deep learning adaptive algorithms, generative adversarial networks and swarm intelligence optimization algorithms to automatically adapt touch area settings and interaction logic for different types of touch devices such as smartphones, tablets and game controllers.

Images