Ancient book repair game framework design method based on configuration

Abstract

The invention provides a configuration-based ancient book repair game framework design method, which comprises the following steps of: analyzing an ancient book damage condition by utilizing image recognition, and generating quantitative description; matching a corresponding virtual repair tool according to the quantitative description, simulating operation through force feedback equipment, and dynamically adjusting the sensitivity and the force of the virtual repair tool; determining a repair scheme according to the quantitative description, optimizing the repair scheme through a machine learning algorithm, and guiding a user to implement the optimized repair scheme through characters and voice; different users are subjected to proficiency grouping, adaptive repair tasks are generated, and the task difficulty is dynamically adjusted; and judging the normalization degree and the validity degree of each user in the process of executing the repair task, and timely providing a correction prompt according to the normalization degree and the validity degree. According to the invention, the digital inheritance of the ancient book repair skill is realized, the immersive learning experience is provided for the user, and the efficiency and quality of ancient book repair talent cultivation are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and particularly to a method for designing a game framework for ancient book restoration based on configuration. Background Art

[0002] With the rapid development of digital technology, the preservation and restoration of ancient books have gradually shifted from traditional manual operations to digital and intelligent directions. However, ancient book restoration is still a process highly dependent on professional knowledge and skills, requiring restorers to have rich experience and excellent techniques. In recent years, gamified learning and working methods have been introduced into the field of ancient book restoration in order to improve the efficiency and quality of ancient book restoration, and at the same time increase the interest and participation in the restoration work.

[0003] Currently, there are already some game-based ancient book restoration applications on the market. These applications allow users to learn and practice restoration techniques in a virtual environment by simulating the process of ancient book restoration. However, most of these applications have some problems that limit their effectiveness in practical applications.

[0004] First of all, most of the existing ancient book restoration games are preset fixed tasks and cannot be adjusted according to the restoration requirements of different ancient books. This results in a single type of game tasks and a lack of diversity, making it difficult to meet the needs of different users for ancient book restoration tasks. At the same time, due to the fixity of the game framework, it is difficult to add new restoration tasks or tools, as well as restoration performance effects, which limits the expandability and playability of the game.

[0005] Secondly, the existing ancient book restoration games lack flexible configuration and expandability. During the game development process, a large amount of customized development of the game framework is often required to adapt to different ancient book restoration tasks. This not only increases the development cost and time, but also limits the generality and adaptability of the game. In addition, due to the closed nature of the game framework, it is difficult for users to add or modify restoration tasks and tools by themselves, reducing the user participation and the interest of the game.

[0006] Furthermore, the existing ancient book restoration games lack personalized settings. Different users may have different interests and needs for ancient book restoration, but the existing games often cannot provide personalized task recommendations and progress management. This leads to users may feel bored during the restoration process, reducing the continuous user participation and learning effect. Summary of the Invention

[0007] The present invention provides a method for designing a game framework for ancient book restoration based on configuration, and the method includes:

[0008] Analyze the damage condition of the ancient book using image recognition to generate a quantitative description;

[0009] Match the corresponding virtual repair tool according to the quantitative description, simulate the operation through a force feedback device, and dynamically adjust the sensitivity and strength of the virtual repair tool;

[0010] Determine the repair plan according to the quantitative description, optimize the repair plan through a machine learning algorithm, and guide the user to implement the optimized repair plan through text and voice;

[0011] Generate adapted repair tasks by grouping users according to proficiency, and dynamically adjust the task difficulty;

[0012] Judge the degree of standardization and effectiveness of each user during the execution of the repair task, and provide corrective prompts immediately according to the degree of standardization and the degree of effectiveness;

[0013] Quantitatively evaluate the repair effect of the ancient book image by analyzing the repaired ancient book image and generate a score;

[0014] Manage and analyze operation data, and continuously optimize the process and task strategy.

[0015] According to an embodiment of the present invention, the use of image recognition to analyze the damage situation of ancient books and generate a quantitative description includes:

[0016] Analyze the input ancient book image by using an image recognition algorithm according to a pre-established ancient book damage feature library to obtain the attribute information of the ancient book, and the attribute information includes material, paper type, and ink characteristics;

[0017] Judge the damage type and degree of damage of the ancient book according to the attribute information of the ancient book, and obtain a quantitative description of the damage situation of the ancient book.

[0018] According to an embodiment of the present invention, the matching of the corresponding virtual repair tool according to the quantitative description, simulating the operation through a force feedback device, and dynamically adjusting the sensitivity and strength of the virtual repair tool includes:

[0019] Match the applicable virtual repair tool from the virtual repair tool library for different quantitative descriptions of the damage situation of ancient books;

[0020] Simulate the operation experience of the virtual repair tool through a force feedback device to obtain the operation data of the user;

[0021] Dynamically adjust the sensitivity and feedback strength of the virtual repair tool by comparing the operation data with the preset operation parameters.

[0022] According to an embodiment of the present invention, the determining of the repair plan according to the quantitative description, optimizing the repair plan through the machine learning algorithm, and guiding the user to implement the optimized repair plan through text and voice includes:

[0023] According to the quantitative description, use knowledge reasoning technology to obtain recommended repair solutions from the knowledge base, and obtain process flow data from the repair solutions;

[0024] Use the decision tree algorithm to analyze and optimize the obtained process flow data to obtain operation step data oriented to key nodes;

[0025] Through natural language processing technology, convert the obtained operation step data into easily understandable text guidance and voice prompt data;

[0026] According to the converted text guidance and voice prompt data, combined with user interaction behavior data, form an interactive guidance feedback mechanism;

[0027] According to the user's interaction behavior data and the preset system configuration file, dynamically load the corresponding virtual repair tools and material data;

[0028] Use real-time rendering technology to present the loaded virtual repair tools and material data to the user.

[0029] According to an embodiment of the present invention, after presenting the loaded virtual repair tools and material data to the user, it further includes:

[0030] Through user behavior analysis technology, obtain the user's repair operation data, compare it with the standard repair process data, judge the accuracy and completion degree of the repair operation, and obtain repair task completion situation data;

[0031] If the difference degree between the repair operation data and the standard process data is greater than the preset difference threshold, trigger the intelligent prompt mechanism to give corresponding correction suggestions and operation guidelines;

[0032] According to the user's repair task completion situation data, combined with the task difficulty and completion quality, use an adaptive algorithm to dynamically adjust the task list and recommendation mechanism to achieve personalized task recommendation and progress management.

[0033] According to an embodiment of the present invention, the generating of adapted repair tasks by grouping different users according to proficiency and dynamically adjusting the task difficulty includes:

[0034] Obtain the user operation data of a large number of users in the ancient book repair game, including the user's repair task selection, tool use, and repair effect evaluation, preprocess and extract features from the user operation data through big data analysis technology to obtain user operation features;

[0035] According to the extracted user operation characteristics, a clustering algorithm is used to group users, obtaining user groups with different proficiency levels, including novice users, intermediate users, and advanced users;

[0036] For user groups with different proficiency levels, corresponding rules for generating repair tasks of appropriate difficulty are designed;

[0037] Combined with the historical operation data and task completion status of different user groups, the matching degree between task difficulty and user proficiency is optimized through a reinforcement learning algorithm, and the task generation rules are dynamically adjusted;

[0038] When a user selects a repair task, a repair task is randomly generated from the corresponding task generation rules according to the user group corresponding to the user's proficiency level, and the corresponding configuration file is loaded, where the configuration file includes the specific steps, required tools, and materials of the repair task;

[0039] The game scene is dynamically loaded and rendered according to the configuration file, providing an interactive repair operation interface;

[0040] During the process of the user executing the repair task, the user's first operation data is recorded in real time. The first operation data includes the tool usage frequency and the repair effect score. Through the analysis and processing of the first operation data, the user's proficiency is dynamically evaluated and updated;

[0041] If the user group corresponding to the user's proficiency level changes, the recalculation of the task difficulty matching degree is triggered to obtain the updated matching degree;

[0042] According to the updated matching degree, the generation rules of subsequent tasks are adjusted to achieve personalized task recommendations;

[0043] When the user completes the repair task and submits the result, the repair effect is evaluated to obtain an evaluation result.

[0044] According to an embodiment of the present invention, judging the degree of standardization and effectiveness of each user during the execution of the repair task, and providing corrective prompts immediately according to the degree of standardization and the degree of effectiveness, includes:

[0045] During the process of the user executing the repair task, the user's second operation data is collected in real time. The second operation data includes the type of tool used, the operation duration, the force magnitude, and the repair part;

[0046] By comparing the second operation data with the standard operation data, the degree of standardization and effectiveness of the user's operation is judged, and corrective feedback prompts are given immediately according to the degree of standardization and the degree of effectiveness.

[0047] According to an embodiment of the present invention, the method of quantifying and evaluating the restoration effect of the ancient book image by analyzing the restored ancient book image and generating a score includes:

[0048] Analyze the restored ancient book image using computer vision technology, and evaluate the restoration effect by comparing it with the original image;

[0049] Evaluate the restoration effect according to the ancient book evaluation indicators to obtain a quantitative score of the restoration effect. The ancient book evaluation indicators include the integrity of the ancient book, the color of the paper, and the clarity of the ink;

[0050] Associate the quantitative score with the operation data of the corresponding user to form a learning sample for optimizing the system.

[0051] According to an embodiment of the present invention, the method further includes:

[0052] Use blockchain technology to record and verify the user's restoration process and results, and realize the automatic acceptance and reward distribution of restoration tasks through the smart contract mechanism.

[0053] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0054] The present invention discloses a method for designing a game framework for ancient book restoration based on configuration. This method analyzes the damage of ancient books through image recognition and matches applicable restoration tools from a virtual tool library. Combining force feedback devices to simulate real restoration experiences, the system generates personalized restoration plans and operation guides according to the attributes and damage levels of ancient books. During the restoration process, the present invention collects user operation data in real time, evaluates the restoration effect, and gives immediate feedback. Continuously optimize the matching of task difficulty and user proficiency through machine learning, and continuously improve the process flow and guiding mechanism using big data analysis. The present invention realizes the digital inheritance of ancient book restoration techniques, provides an immersive learning experience for users, and effectively improves the efficiency and quality of the cultivation of ancient book restoration talents. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0056] Figure 1 is a flowchart of a method for designing a game framework for ancient book restoration based on configuration of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0058] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] As Figure 1 , this application proposes a method for designing a configuration-based ancient book restoration game framework, including:

[0060] S101. Use image recognition to analyze the damage condition of the ancient book and generate a quantitative description;

[0061] S102. Match the corresponding virtual restoration tool according to the quantitative description, simulate the operation through a force feedback device, and dynamically adjust the sensitivity and strength of the virtual restoration tool;

[0062] S103. Determine the restoration plan according to the quantitative description, optimize the restoration plan through a machine learning algorithm, and guide the user to implement the optimized restoration plan through text and voice;

[0063] S104. Generate adapted restoration tasks by grouping different users according to proficiency and dynamically adjust the task difficulty;

[0064] S105. Judge the degree of standardization and effectiveness of each user during the execution of the restoration task, and provide corrective prompts immediately according to the degree of standardization and the degree of effectiveness;

[0065] S106. Quantitatively evaluate the restoration effect of the ancient book image by analyzing the restored ancient book image and generate a score;

[0066] S107. Manage and analyze operation data, and continuously optimize the process and task strategy.

[0067] In some embodiments, step S101 specifically includes:

[0068] According to a pre-established ancient book damage feature library, use an image recognition algorithm to analyze the input ancient book image to obtain the attribute information of the ancient book, and the attribute information includes material, paper type, and ink characteristics;

[0069] According to the attribute information of the ancient book, judge the damage type and damage degree of the ancient book to obtain a quantitative description of the ancient book damage condition.

[0070] In this embodiment, according to the pre-established ancient book damage feature library, an image recognition algorithm is used to analyze the input ancient book image to obtain attributes such as the material of the ancient book, the type of paper, and the characteristics of the ink marks. By analyzing the attributes of the ancient book, the damage type of the ancient book, such as insect damage, breakage, fading, etc., is judged, and the degree of damage is quantitatively evaluated. According to the attributes such as the material of the ancient book, the type of paper, and the characteristics of the ink marks, as well as the damage type and degree, a suitable repair plan is matched from the repair tool and material database. If there are multiple applicable repair plans, a comprehensive evaluation is carried out according to factors such as the degree of damage, the repair effect, and the time consumption to determine the optimal repair plan. The selected repair plan is converted into a series of repair steps and operations to generate the task flow for ancient book repair. According to the task flow, the corresponding virtual repair tools and materials are dynamically loaded and presented on the game interface for players to operate. A physical engine is used to simulate the tool operation and material changes during the repair process and render the repair effect in real time. By comparing the ancient book images before and after repair, the repair effect is evaluated, and the player's score is calculated according to indicators such as the repair quality and the completion time. The player's repair result data is uploaded to the server for updating the ancient book damage feature library and the repair plan database, and the task difficulty and reward mechanism are adjusted according to the player's performance.

[0071] Specifically, first, a convolutional neural network is used to extract features from ancient book images. By comparing with a pre-established feature library containing 100,000 damaged ancient book images, the cosine similarity algorithm is used to calculate the distance between feature vectors. If the similarity is greater than 8, it is determined to be the same type of damage. Then, based on the image segmentation algorithm, the damaged area is extracted, and the proportion of the damaged area is quantitatively evaluated by counting pixel points. The damage degree can be divided into three levels: mild (less than 10%), moderate (10% - 30%), and severe (greater than 30%). After determining the damage type and degree, a suitable solution is matched from a database containing 500 repair solutions, and the decision tree algorithm is used to evaluate the applicability of the solution. Considering factors such as damage degree, repair effect, and time consumption, the optimal solution is selected by weighted scoring. Next, the repair solution is decomposed into a series of specific steps, and the preconditions and subsequent processes of each step are defined by a finite state machine to form a complete task flow chart. In the game engine, 3D models and texture resources are dynamically loaded according to the task flow, and the physical engine is used to simulate the property parameters of tools and materials, such as mass, elastic coefficient, friction coefficient, etc. By solving Newton's equations of motion, their force changes and motion trajectories are calculated in real time, and a realistic repair process is rendered. Finally, by using the image difference algorithm to compare the ancient book images before and after repair, the repair ratio of the damaged area is calculated. If the ratio is greater than 95%, it is determined that the repair is successful. At the same time, the completion time of the player is recorded, the score of this task is calculated according to the preset scoring criteria, and the result data is uploaded to the server. The server aggregates the repair data of all players, continuously optimizes the damaged feature library and repair solution library using machine learning algorithms, and at the same time dynamically adjusts the task difficulty and reward intensity by clustering the ability levels of players to ensure the playability and challenge of the game.

[0072] In some embodiments, step S102 specifically includes:

[0073] For the quantitative description of different ancient book damage situations, a suitable virtual repair tool is matched from the virtual repair tool library;

[0074] The operation experience of the virtual repair tool is simulated through a force feedback device to obtain the operation data of the user;

[0075] By comparing the operation data with preset operation parameters, the sensitivity and feedback strength of the virtual repair tool are dynamically adjusted.

[0076] In this embodiment, according to the damage situation and characteristics of the ancient book, the applicable virtual repair tool combination is matched from the repair tool configuration file to obtain targeted repair tool parameters. The ancient book image is analyzed by computer vision technology to identify different types of damaged areas such as insect bites and breakages, and obtain damage feature data. If the damage type is insect bites, a hole filling tool is selected; if the damage type is breakage, a mounting tool is selected; the parameters of the selected tool are dynamically adjusted according to the identified damage characteristics. The virtual repair tool is manipulated by a force feedback device to obtain the user's operation data, including the force size, action time, etc. The user operation data is compared with the preset standard operation parameters, the operation deviation value is calculated, and the user is judged. Whether the user's operation is standardized If the operation deviation value exceeds the set threshold, the sensitivity and feedback strength of the virtual repair tool are dynamically adjusted to guide the user to perform standardized operations Based on the user's operation process data, a preview of the repair effect of the ancient book is generated in real time, and compared with the original ancient book image, and the similarity of the repair effect is calculated If the similarity of the repair effect does not reach the preset target value, it returns to step 4 and prompts the user to continue the repair; if the similarity of the repair effect meets the requirements, the repair task of the current tool is completed Comprehensively analyze the data of all repair tool operations completed by the user, evaluate the overall repair quality, feedback the evaluation results to the user, and store the repair operation data in the user profile for subsequent task recommendations and dynamic optimization of repair effects

[0077] Specifically, first, the system will match the optimal tool combination from the pre-set repair tool configuration file according to the characteristics of the damaged ancient books, such as the proportion of the area damaged by insects, the length of the breakage, etc. For example, if the proportion of the area damaged by insects exceeds 20%, a large-area hole-filling tool is required, and if the length of the breakage exceeds 10 cm, a reinforced pasting tool is required. Then, it uses image segmentation algorithms such as U-Net to analyze the high-definition photos of ancient books, identifies different types of damaged areas through a deep learning model, and extracts feature parameters such as their positions, sizes, and shapes. According to the recognition results, it dynamically adjusts the parameters of the selected tools, such as the filling material of the hole-filling tool and the bonding strength of the pasting tool. When the user performs virtual repair operations using a force feedback device, the system collects pressure sensor data at a frequency of 1000 Hz, calculates the magnitude and direction of the user's applied force. At the same time, it refers to the standard hand movement parameters in the ergonomics database and calculates the deviation value between the user's actual operation and the standard operation through the DTW algorithm. If the deviation value exceeds 20%, the feedback force of the virtual tool is increased by 50% and the sensitivity is reduced to guide the user to operate in a standard manner. The system will also simulate the deformation and attachment process of the repair material based on the user's real-time operation data using a physics engine to generate a realistic preview of the ancient book repair effect. Through image similarity algorithms such as SSIM, it evaluates the closeness of the virtual repair result to the target state, and if the similarity reaches more than 95%, it is determined that the repair is completed. Finally, the system comprehensively analyzes data such as the time and accuracy of the user to complete a set of repair processes, gives a comprehensive score of the repair quality through a weighted average algorithm, and stores the score and operation data in the user profile for intelligent recommendation of subsequent suitable repair tasks and optimization of tool and process settings.

[0078] In some embodiments, step S103 specifically includes:

[0079] According to the quantitative description, use knowledge reasoning technology to obtain a recommended repair plan from the knowledge base and obtain process flow data from the repair plan;

[0080] Use a decision tree algorithm to analyze and optimize the obtained process flow data to obtain operation step data oriented to key nodes;

[0081] Through natural language processing technology, convert the obtained operation step data into easy-to-understand text guidelines and voice prompt data;

[0082] According to the converted text guidelines and voice prompt data, combined with user interaction behavior data, form an interactive guidance feedback mechanism;

[0083] According to the user's interaction behavior data and the pre-set system configuration file, dynamically load the corresponding virtual repair tools and material data;

[0084] Using real-time rendering technology, the loaded virtual restoration tools and material data are presented to the user.

[0085] Further, after presenting the loaded virtual restoration tools and material data to the user, it further includes:

[0086] Through user behavior analysis technology, obtain the user's restoration operation data, compare it with the standard restoration process data, judge the accuracy and completion degree of the restoration operation, and obtain the restoration task completion situation data;

[0087] If the difference degree between the restoration operation data and the standard process data is greater than the preset difference threshold, trigger the intelligent prompt mechanism to give corresponding corrective suggestions and operation guidelines;

[0088] According to the user's restoration task completion situation data, combined with the task difficulty and completion quality, adopt an adaptive algorithm to dynamically adjust the task list and recommendation mechanism to achieve personalized task recommendation and progress management.

[0089] In this embodiment, according to the attribute information and damage situation data of the ancient books, use knowledge reasoning technology to obtain the recommended restoration plan and corresponding process flow data from the knowledge base. Use the decision tree algorithm to analyze and optimize the obtained process flow data to obtain the operation step data guided by key nodes. Through natural language processing technology, convert the obtained operation step data into easy-to-understand text guidelines and voice prompt data. According to the converted text guidelines and voice prompt data, combined with the user interaction behavior data, form an interactive guidance feedback mechanism. According to the user's restoration operation data and the system configuration file, dynamically load the corresponding virtual restoration tools and material data. Using real-time rendering technology, present the loaded virtual restoration tools and material data to the user to provide an immersive restoration experience. Through user behavior analysis technology, obtain the user's restoration operation data and compare it with the standard restoration process data to judge the accuracy and completion degree of the restoration operation. If the difference between the restoration operation data and the standard process data is large, trigger the intelligent prompt mechanism to give corresponding corrective suggestions and operation guidelines. According to the user's restoration task completion situation data, combined with factors such as task difficulty and completion quality, adopt an adaptive algorithm to dynamically adjust the task list and recommendation mechanism to achieve personalized task recommendation and progress management.

[0090] In some embodiments, step S104 specifically includes:

[0091] Obtain the user operation data of a large number of users in the ancient book restoration game, including the user's restoration task selection, tool use, and restoration effect evaluation. Through big data analysis technology, preprocess and extract features from the user operation data to obtain user operation features;

[0092] According to the extracted user operation characteristics, a clustering algorithm is used to group users, obtaining user groups with different proficiency levels, including novice users, intermediate users, and advanced users;

[0093] For user groups with different proficiency levels, corresponding rules for generating repair tasks with appropriate difficulty levels are designed;

[0094] Combined with the historical operation data and task completion status of different user groups, the matching degree between task difficulty and user proficiency is optimized through a reinforcement learning algorithm, and the task generation rules are dynamically adjusted;

[0095] When a user selects a repair task, a repair task is randomly generated from the corresponding task generation rules according to the user group corresponding to the user's proficiency level, and the corresponding configuration file is loaded. The configuration file includes the specific steps, required tools, and materials of the repair task;

[0096] The game scene is dynamically loaded and rendered according to the configuration file, providing an interactive repair operation interface;

[0097] During the process of the user executing the repair task, the user's first operation data is recorded in real time. The first operation data includes the tool usage frequency and the repair effect score. By analyzing and processing the first operation data, the user's proficiency is dynamically evaluated and updated;

[0098] If the user group corresponding to the user's proficiency level changes, the recalculation of the task difficulty matching degree is triggered to obtain the updated matching degree;

[0099] According to the updated matching degree, the generation rules of subsequent tasks are adjusted to achieve personalized task recommendations;

[0100] When the user completes the repair task and submits the result, the repair effect is evaluated to obtain the evaluation result.

[0101] In this embodiment, a large amount of operation data of users in the ancient book restoration game is obtained, including information such as users' restoration task selection, tool use, and restoration effect evaluation. The data is preprocessed and feature extracted through big data analysis techniques. According to the extracted user operation features, a clustering algorithm is used to group users, obtaining user groups with different proficiency levels, such as novice users, intermediate users, and advanced users. For user groups with different proficiency levels, corresponding difficulty rules for generating restoration tasks are designed. For example, the tasks for novice users have lower difficulty and detailed operation steps, while the tasks for advanced users have higher difficulty and concise operation steps. Through a reinforcement learning algorithm, combined with the user's historical operation data and task completion situation, the matching degree between task difficulty and user proficiency is continuously optimized, and the task generation rules are dynamically adjusted. When a user selects a restoration task, a restoration task is randomly generated from the corresponding task generation rules according to the user's proficiency level, and the corresponding configuration file is loaded. The configuration file contains information such as the specific steps of the restoration task, required tools, and materials. The system dynamically loads and renders the game scene according to the configuration file, providing an interactive restoration operation interface. During the process of the user executing the restoration task, the user's operation data, such as tool use frequency and restoration effect score, is recorded in real time. Through data analysis and machine learning techniques, the user's proficiency is dynamically evaluated and updated. If the user's proficiency changes, it triggers a recalculation of the task difficulty matching degree. According to the updated matching degree, the subsequent task generation rules are adjusted to achieve personalized task recommendation. When the user completes the restoration task and submits the result, the system evaluates the restoration effect and stores the evaluation result together with the user operation data for subsequent user grouping, task generation rule optimization, etc., forming a data-driven closed-loop feedback mechanism.

[0102] In some embodiments, step S105 specifically includes:

[0103] During the process of the user executing the restoration task, the second operation data of the user is collected in real time, and the second operation data includes the tool type used, operation duration, force magnitude, and restoration part;

[0104] By comparing the second operation data with the standard operation data, the degree of standardization and effectiveness of the user's operation is judged, and a corrective feedback prompt is given immediately according to the degree of standardization and the degree of effectiveness.

[0105] In this embodiment, according to the repair task selected by the user, the corresponding standard operation data configuration file is obtained, and the evaluation indicators and thresholds are determined. The operation data of the user during the repair process is collected in real time by sensors, including the type of tool used, operation duration, force magnitude, repair location, etc. Through data preprocessing, the collected original operation data is cleaned, normalized, etc. to obtain standardized operation data. The standardized operation data is compared with the standard operation data, and the deviation value is calculated. If the deviation value exceeds the threshold range, it is determined that there is a problem with the user's operation, and the feedback prompt mechanism is triggered. According to the magnitude and direction of the deviation value, the type and content of the feedback prompt are determined, and corrective guidance information is generated. Using technical means such as virtual reality and animation, the feedback prompt information is presented to the user in real time to guide the user to make operation adjustments. Continuously track the change of the user's operation data, judge whether the user has made operation improvements according to the feedback prompt, and update the feedback strategy accordingly. When the user completes the repair task, comprehensively analyze the operation data of the whole process, evaluate the repair effect and standardization of the user, and record the results in the user profile for subsequent task recommendation and ability assessment.

[0106] In some embodiments, step S106 specifically includes:

[0107] The repaired ancient book image is analyzed using computer vision technology, and the repair effect is evaluated by comparing it with the original image;

[0108] The repair effect is evaluated according to the ancient book evaluation indicators to obtain a quantitative score of the repair effect. The ancient book evaluation indicators include the integrity of the ancient book, the color of the paper, and the clarity of the ink;

[0109] The quantitative score is associated with the operation data of the corresponding user to form a learning sample for optimizing the system.

[0110] In this embodiment, computer vision technology is used to preprocess the restored ancient book images, including operations such as image denoising, enhancement, and segmentation, to obtain image data suitable for analysis. Through feature extraction algorithms, key features reflecting indicators such as the integrity of ancient books, the color of paper, and the clarity of ink marks are extracted from the preprocessed restored images. The original ancient book images are obtained, and also preprocessed and feature extracted to obtain the feature data of the original images. The features of the restored images are compared with the features of the original images, and similarity calculation methods are used to quantitatively evaluate the restoration effect. According to the feature comparison results, considering various indicators comprehensively, through weighted average or other scoring functions, a quantitative score of the restoration effect is calculated. The operation data of the user during the restoration process, such as the tools used, restoration steps, time, etc., are obtained and associated with the restoration effect score to form learning samples. Machine learning algorithms, such as support vector machines, decision trees, etc., are used to train the restoration effect evaluation model, and the scoring function and weights are continuously optimized. When subsequent users perform ancient book restoration, the trained evaluation model is used to evaluate the restoration effect in real time, and the task difficulty is dynamically adjusted according to the scoring results and intelligent prompts are provided. Continuously collect the restoration data and scoring feedback of users, and retrain and optimize the evaluation model regularly to improve the intelligent level and user experience of the system.

[0111] In some embodiments, the above steps S101 to S107 further include: using blockchain technology to record and verify the restoration process and results of users, and realizing the automatic acceptance and reward distribution of restoration tasks through smart contract mechanisms.

[0112] In this embodiment, blockchain technology is used to record and verify the restoration process and results of users, and the automatic acceptance and reward distribution of restoration tasks are realized through smart contract mechanisms, motivating users to continuously participate and contribute.

[0113] The above-described embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for designing a configuration-based ancient book restoration game framework, characterized in that The method includes: Analyze the damage condition of ancient books using image recognition to generate a quantitative description; Match the corresponding virtual repair tool according to the quantitative description, simulate the operation through a force feedback device, and dynamically adjust the sensitivity and strength of the virtual repair tool; Determine the repair plan according to the quantitative description, optimize the repair plan through a machine learning algorithm, and guide the user to implement the optimized repair plan through text and voice; Generate adapted repair tasks by grouping different users according to their proficiency, and dynamically adjust the task difficulty; Judge the degree of standardization and effectiveness of each user during the execution of the repair task, and provide corrective prompts immediately according to the degree of standardization and the degree of effectiveness; Quantitatively evaluate the repair effect of the ancient book image by analyzing the repaired ancient book image and generate a score; Manage and analyze operation data, and continuously optimize the process and task strategy.

2. The method according to claim 1, wherein The step of analyzing the damage condition of ancient books using image recognition to generate a quantitative description includes: According to a pre-established ancient book damage feature library, use an image recognition algorithm to analyze the input ancient book image to obtain the attribute information of the ancient book, and the attribute information includes material, paper type, and ink characteristics; Judge the damage type and degree of the ancient book according to the attribute information of the ancient book to obtain a quantitative description of the ancient book damage condition.

3. The method according to claim 1, characterized in that, The step of matching the corresponding virtual repair tool according to the quantitative description, simulating the operation through a force feedback device, and dynamically adjusting the sensitivity and strength of the virtual repair tool includes: Match the applicable virtual repair tool from the virtual repair tool library according to the quantitative description of different ancient book damage conditions; Simulate the operation experience of the virtual repair tool through a force feedback device to obtain the operation data of the user; Dynamically adjust the sensitivity and feedback strength of the virtual repair tool by comparing the operation data with preset operation parameters.

4. The method according to claim 1, wherein The step of determining the repair plan according to the quantitative description, optimizing the repair plan through the machine learning algorithm, and guiding the user to implement the optimized repair plan through text and voice includes: According to the quantitative description, use knowledge reasoning technology to obtain the recommended repair plan from the knowledge base, and obtain the process flow data from the repair plan; Use a decision tree algorithm to analyze and optimize the obtained process flow data to obtain operation step data oriented to key nodes; Through natural language processing technology, convert the obtained operation step data into easy-to-understand text guidelines and voice prompt data; According to the converted text guidelines and voice prompt data, combined with the user interaction behavior data, form an interactive guidance feedback mechanism; Dynamically load the corresponding virtual repair tool and material data according to the user's interaction behavior data and a preset system configuration file; Use real-time rendering technology to present the loaded virtual repair tool and material data to the user.

5. The method according to claim 4, characterized in that, After presenting the loaded virtual repair tool and material data to the user, it further includes: Obtain the repair operation data of the user through user behavior analysis technology, compare it with the standard repair process data, judge the accuracy and completion degree of the repair operation, and obtain the repair task completion situation data; If the difference degree between the repair operation data and the standard process data is greater than the preset difference threshold, trigger the intelligent prompt mechanism to give corresponding corrective suggestions and operation guidelines; According to the repair task completion data of the user, combined with the task difficulty and completion quality, use an adaptive algorithm to dynamically adjust the task list and recommendation mechanism to achieve personalized task recommendation and progress management.

6. The method according to claim 1, characterized in that The dynamically adjusting the task difficulty by grouping different users according to proficiency and generating adapted repair tasks includes: Obtain the user operation data of a large number of users in the ancient book repair game, including the user's repair task selection, tool use, and repair effect evaluation. Preprocess and extract features from the user operation data through big data analysis technology to obtain user operation features; According to the extracted user operation features, use a clustering algorithm to group users to obtain user groups with different proficiencies, and the user groups include novice users, intermediate users, and advanced users; For user groups with different proficiencies, design corresponding repair task generation rules; Combined with the historical operation data and task completion of different user groups, optimize the matching degree between task difficulty and user proficiency through a reinforcement learning algorithm, and dynamically adjust the task generation rules; When the user selects a repair task, randomly generate a repair task from the corresponding task generation rules according to the user group corresponding to the user's proficiency level, and load the corresponding configuration file, and the configuration file includes the specific steps, required tools, and materials of the repair task; Dynamically load and render the game scene according to the configuration file to provide an interactive repair operation interface; During the process of the user executing the repair task, record the user's first operation data in real time, and the first operation data includes the tool use frequency and the repair effect score. Analyze and process the first operation data to dynamically evaluate and update the user's proficiency; If the user group corresponding to the user's proficiency level changes, trigger the recalculation of the task difficulty matching degree to obtain the updated matching degree; According to the updated matching degree, adjust the generation rules of subsequent tasks to achieve personalized task recommendation; When the user completes the repair task and submits the result, evaluate the repair effect to obtain the evaluation result.

7. The method according to claim 1, characterized in that The judging the degree of standardization and effectiveness of each user during the execution of the repair task and immediately providing corrective prompts according to the degree of standardization and the degree of effectiveness includes: During the process of the user executing the repair task, collect the user's second operation data in real time, and the second operation data includes the tool type used, operation duration, force magnitude, and repair location; By comparing the second operation data with the standard operation data, judge the degree of standardization and effectiveness of the user's operation, and immediately give a corrective feedback prompt according to the degree of standardization and the degree of effectiveness.

8. The method according to claim 1, characterized in that The quantitatively evaluating the repair effect of the ancient book image and generating a score by analyzing the repaired ancient book image includes: Use computer vision technology to analyze the repaired ancient book image and evaluate the repair effect by comparing it with the original image; Evaluate the restoration effect according to the ancient book evaluation indicators to obtain a quantitative score of the restoration effect. The ancient book evaluation indicators include the integrity of the ancient book, the color of the paper, and the clarity of the ink marks; Associate the quantitative score with the operation data of the corresponding user to form a learning sample for optimizing the system.

9. The method according to claim 1, characterized in that The method further includes: Adopt blockchain technology to record and verify the user's restoration process and results, and realize the automatic acceptance and reward distribution of restoration tasks through the smart contract mechanism.

Images