Studio video recording and playing system

Abstract

The invention discloses a studio video recording and playing system, belongs to the technical field of computer application, and aims to solve the problems that service rule-data splitting, rehearsal depends on actual equipment, and abnormity post-event processing is carried out. Comprising the steps that a rule data system construction module generates a rule-data association group through three-dimensional rule classification, three-level data attribute marking and a three-source input association mechanism; the twinborn rehearsal scheme optimization module depends on a digital twinborn environment and a business rule-data association group reference library to realize full-process undisturbed rehearsal and optimal adaptive configuration screening; the recording and broadcasting real-time dynamic optimization module completes risk beforehand early warning and accurate adjustment through time sequence-multi-dimensional index double-matrix monitoring, a preorder pre-judgment window and abnormal feature quantitative matching; according to the method, demand-rule-resource dynamic matching is realized, configuration conflicts and resource insufficiency risks are checked in advance, the stability and high quality of the whole recording and broadcasting process are ensured, and the process normalization and the resource utilization rate are improved.

Description

Technical Field

[0001] This invention belongs to the field of computer application technology, and specifically relates to a studio video recording and playback system. Background Technology

[0002] With the diversification of studio recording scenarios, the requirements for process accuracy and operational stability are increasing. However, existing technologies have shortcomings: chaotic and insufficiently correlated rule and data management; lack of non-intrusive rehearsal and quantitative adaptation mechanisms; and single monitoring methods without pre-judgment and precise anomaly handling solutions. Therefore, this invention urgently needs to solve the following technical problems: In studio recording, business rules, equipment configuration and real-time data are disconnected, making it difficult to dynamically match requirements, rules and resources; when switching scenes or moving equipment, manual trial and error parameter tuning is required, which cannot predict configuration conflicts or automatically reselect links, resulting in problems such as incorrect camera blocking and inefficient resource configuration. Pre-recording debugging relies on actual equipment, making it impossible to simulate the entire process without interference to identify risks in advance. When the configuration and adaptation are not up to standard, there is a lack of quantitative screening criteria. Adaptation adjustments rely on blind trial and error, which is inefficient and prone to leaving hidden dangers. The monitoring indicators for recording and broadcasting are too simplistic and lack the ability to predict risks in advance. Anomalies are mostly discovered after the fact, and there is no targeted anomaly handling and matching mechanism. Adjustments can easily disrupt existing processes and cause secondary risks. Therefore, we propose a studio video recording and playback system. Summary of the Invention

[0003] The purpose of this invention is to provide a studio video recording and playback system to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a studio video recording and playback system, comprising: Rule data system construction module: acquire business rules for recording and playback scenarios, build a business rule index library; collect full data of the recording and playback chain, build a basic data view of the recording and playback chain; build a rule attribute mapping library, establish a three-source input channel and decision association mechanism, and generate rule-data association groups that match decision requirements; Twin rehearsal optimization module: Construct a digital twin environment for live physical system rehearsal, clarify the business rule requirements of each time node based on the time-series execution baseline, analyze the rule-data association group, and determine whether the requirement adaptability is qualified; if not qualified, rematch the adaptable three-level link attribution attribute; integrate the rule-data association group of all time-series nodes, and output the optimized live execution plan; Recording and playback real-time dynamic optimization module: Constructs a theoretical standard matrix of time-series and multi-dimensional indicators, and simultaneously constructs an actual monitoring matrix of time-series and multi-dimensional indicators; analyzes the execution adaptation prediction value of the business rules of the time-series nodes to be executed, and judges whether there is any risk of execution anomalies; if there is an anomaly, a solution is matched, the adaptation configuration is adjusted, until the full-process recording and playback monitoring optimization is completed.

[0005] Preferably, the construction process of the business rule index library, the basic data view of the recording and playback link, and the rule attribute mapping library is as follows: Obtain the business rules corresponding to the current recording and playback scenario, classify and organize them according to three dimensions: scenario allocation type, preset rule priority, and constraint object, build a business rule index library, and assign a unique index identifier to each type of business rule; Real-time collection of all basic data of each component of the recording and playback link, marking each data item with the corresponding three-level link affiliation attribute, and classifying them according to the attribute and collection timestamp to form a structured basic data view of the recording and playback link; Construct a rule attribute mapping library, which pre-sets and stores the three-level link attribution attributes corresponding to each type of business rule in the business rule index library.

[0006] Preferably, the specific process of establishing a three-source input channel and decision association mechanism to generate rule-data association groups that match decision requirements is as follows: The business rule index library, rule attribute mapping library, and basic data view of recording and playback link are used as the three sources of input for the intelligent decision-making terminal, and the rule calling channel, mapping matching channel and data calling channel are constructed simultaneously. The rule invocation channel invokes the target rule through the decision requirement-index identifier matching logic. The mapping matching channel extracts the three-level link attribution attribute corresponding to the target rule. The data invocation channel reads the standardized data corresponding to this attribute. A decision association triggering mechanism is constructed, which follows the business decision requirement → matching business rules → extracting mapping data attributes → invoking the target data link, binding the target business rule with the corresponding data, and generating a rule-data association group.

[0007] Preferably, the process of constructing a digital twin environment for live physical system rehearsal, clarifying the business rule requirements for each time node based on the time-series execution baseline, and analyzing the specific process of rule-data association groups is as follows: A digital mirror image corresponding to the studio's live broadcast physical system is constructed based on digital twin technology to form a digital twin environment for rehearsals; Obtain the execution baseline in time sequence, transform the execution actions into business rule execution requirements with timestamps according to the time sequence, and clarify the corresponding time sequence nodes; start the pre-rehearsal environment, input the requirements into the built-in decision association trigger mechanism in sequence, which calls the integrated business rule index library, rule attribute mapping library and the basic data view of the recording and playback link in the digital image, and generates a rule-data association group with time sequence identifier through matching rules → extracting attributes → calling data links.

[0008] Preferably, the specific process for determining whether the business rule requirements are suitable is as follows: Construct a benchmark library for business rules and data association groups. The library stores the standard three-tier link attribution attributes, standard resource parameter security ranges, and alternative three-tier link attribution attribute sets corresponding to each type of business rule. After the rule-data association group with time sequence identifier is generated, the benchmark library is searched with the target business rule to match the corresponding standard attributes and security intervals. The three-level link attribution attributes and target resource data in the association group are compared with them respectively. If the attributes are completely consistent and the data is within the security interval, the adaptability is deemed qualified, the resource data and timestamp corresponding to the digital image are updated, and the requirements of the next time sequence node are processed; otherwise, the adaptability is deemed unqualified.

[0009] If the preferred option is not found, then a new matching of the three-tier link attribution attributes will be performed. When the compatibility is not up to standard, the business rule-data association group benchmark library is retrieved based on the current target business rule, the corresponding set of alternative three-level link attribution attributes is retrieved, each alternative three-level link attribution attribute is read in turn and the corresponding resource data in the digital image is obtained, and the standard three-level link attribution attribute and standard resource parameter security range corresponding to the business rule in the business rule-data association group benchmark library are matched at the same time. By combining the parameter values ​​of core performance parameters with preset weight coefficients, the matching degree between the attribution attributes of each candidate third-level link and the attribution attributes of the standard third-level link is calculated. The candidate third-level link attribution attributes with the highest matching degree are selected, and their corresponding resource data is compared with the standard resource parameter security range in the benchmark library of business rules-data association group. If the data is within the safe range, replace the original three-level link attribution attribute in the rule-data association group with the alternative three-level link attribution attribute, regenerate the rule-data association group with time sequence identifier, update the digital mirror resource data and timestamp, and continue to process the next time sequence node requirement; If the data is not within the safe range, remove the candidate third-level link attribution attribute and repeat the screening and comparison until a suitable candidate third-level link attribution attribute is found and the update is completed.

[0010] Preferably, the specific process of integrating the rules-data association groups of all time-series nodes and outputting the optimized live streaming execution plan is as follows: Once all business rule execution requirements corresponding to all time-series nodes in the time-series execution baseline have completed the above matching, comparison, and processing operations, the final rule-data association group corresponding to all time-series nodes is traversed. The target business rule, the finally determined three-level link attribution attribute, and the updated resource data of each node are extracted in the order of the time-series nodes. The extracted information is integrated into structured data, and the structured data is output as the optimized live broadcast execution scheme.

[0011] Preferably, the specific process of analyzing the execution adaptation prediction value of the business rules of the time sequence node to be executed and determining whether there is a risk of execution anomalies is as follows: Based on the optimized live streaming execution scheme, we extract the theoretical values ​​of multi-dimensional core indicators of devices and transmission links corresponding to the three-level link affiliation attributes associated with each time-series node, and construct a time-series-multi-dimensional indicator theoretical standard matrix. During the actual recording and broadcasting, multi-dimensional core indicator operation data of the executed nodes are collected in real time, and a time-series-multi-dimensional indicator actual monitoring matrix is ​​constructed simultaneously. Pre-set a preceding prediction window for the business rules of the next time sequence node, extract the consistency coefficient between actual and theoretical data, the average deviation of indicators, and the relative fluctuation range of each type of core indicator in the window, and calculate the core indicator adaptation score after normalization and combining it with the preset weight coefficient. Then, summarize the scores and combine them with the corresponding preset weight coefficient to obtain the execution adaptation prediction value. If the predicted value is not lower than the preset suitable qualification threshold, it is determined that the conditions for normal execution are met, and the plan is executed and the relevant matrices and data are updated; if it is lower, it is determined that there is a risk of execution abnormality.

[0012] Preferably, if an anomaly occurs, a solution is matched, and the adaptation configuration is adjusted until the entire recording and playback monitoring optimization process is completed. The specific process is as follows: When an execution anomaly risk is determined, the multi-dimensional core comparison parameters in the preceding prediction window corresponding to the business rule are extracted, and an anomaly feature set is constructed. The low-dimensional anomaly feature vector is obtained by principal component analysis and three-dimensional dimensionality reduction. The vector is then substituted into the anomaly feature-solution mapping library, and the target solution with the highest fit is selected by feature vector cosine similarity matching and sent to the operation and maintenance management terminal. Operators simulate and verify the target solution in the digital mirror, adjust the three-level link attribution attribute or resource configuration parameters corresponding to the business rules, regenerate the rule-data association group and correct the theoretical values ​​of the core indicators related to the time series-multi-dimensional indicator theoretical standard matrix, repeat the prediction process until the prediction value meets the standard, and then execute the business rules of the next time series node. Repeat the above process until all time-series node business rules have been executed, thus completing the optimization of the entire recording and broadcasting monitoring process.

[0013] Compared with the prior art, the beneficial effects of the present invention are: (1) This studio video recording and playback system builds a business rule index library, a basic data view of the recording and playback link and a rule attribute mapping library, and establishes a three-source input channel and decision association mechanism. It classifies business rules according to three dimensions, marks data according to three-level attributes and establishes accurate correspondence, breaks down the barriers between the three, realizes dynamic matching of "demand-rules-resources", eliminates the need for manual trial and error parameter adjustment, solves the problems of camera position mis-shielding and inefficient resource allocation, and improves the standardization of the recording and playback process and the utilization rate of resources.

[0014] (2) This studio video recording and playback system constructs a digital twin pre-playing environment that corresponds to the physical system in a 1:1 manner. It is equipped with a business rule-data association group benchmark library to achieve uninterrupted simulation of the entire recording and playback process and time sequence node resource-parameter consistency verification. Then, it uses core performance parameters to quantify the matching degree and select the optimal candidate three-level link attribution attributes. It does not rely on actual equipment debugging, and it can check configuration conflicts and resource shortage risks in advance, avoid blind trial and error and leave hidden dangers, and reduce the probability of live broadcast lag and interruption.

[0015] (3) This studio video recording and playback system, by constructing a time-series-multi-dimensional indicator dual matrix monitoring system and a pre-prediction window, realizes early warning of abnormal risks. Then, through abnormal feature dimensionality reduction processing and abnormal feature-solution mapping library quantitative matching, combined with digital mirror simulation to verify the adjustment effect, it avoids the drawbacks of single indicator monitoring, post-event alarm and blind "blind adjustment", realizes accurate measures for abnormalities, reduces the secondary risks caused by adjustment, shortens the fault recovery time, and ensures stable and high-quality recording and playback throughout the entire process. Attached Figure Description

[0016] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1; Please see Figure 1 The present invention provides a studio video recording and playback system, comprising: a rule data system construction module, a twin pre-play scheme optimization module, and a real-time dynamic optimization module for recording and playback; The rule data system construction module includes: acquiring business rules for recording and playback scenarios and building a business rule index library; collecting full data from the recording and playback chain and building a basic data view of the recording and playback chain; building a rule attribute mapping library; establishing a three-source input channel and decision association mechanism; and generating rule-data association groups that match decision requirements. The specific process is as follows: Obtain the business rules corresponding to the current recording and playback scenario, including but not limited to: prioritizing celebrity close-ups, disabling overloaded devices, shot switching timing specifications, and redundancy backup strategies. The acquired business rules are categorized and organized according to three dimensions: scenario type, preset rule priority, and constraint object. A business rule index library is built based on the categorization and organization results, and a unique index identifier is assigned to each type of business rule. The system acquires all basic data of each component of the current recording and playback link in real time, marks each basic data item with its corresponding three-level link affiliation attribute (i.e., link component - specific device identifier - parameter type), and performs preprocessing. The complete set of basic data includes: camera configuration information, basic equipment parameters, encoding standards, and material storage paths, etc. The recording and playback link consists of: camera equipment, encoding equipment, and material storage units, etc. The preprocessed full set of basic data is classified and organized according to the three-level link attribution attributes and collection timestamps corresponding to each basic data item, forming a structured recording and playback link basic data view (for example, when the collection timestamp is T1, the data with the attribution attribute of "camera position device-camera position 1-configuration parameters" includes the resolution, frame rate, and other information of the high-definition camera at camera position 1). Construct a rule attribute mapping rule library. In this library, for each type of business rule in the business rule index library, preset and store its corresponding three-level link affiliation attribute (e.g., business rule R001 (priority star close-up) → mapping "camera device - camera position 1 (dedicated camera position for star close-up) - configuration parameters" "camera device - camera position 1 - load parameters"). The business rule index library, rule attribute mapping rule library, and recording and playback link basic data view are used as the three sources of input for the intelligent decision-making end (based on: business rules providing constraints, mapping rules providing association basis, and data view providing supporting data, the three work together to achieve accurate matching), and at the same time, rule calling channel, mapping matching channel, and data calling channel are constructed. Rule invocation channel: Configure decision requirements - index identifier matching logic, quickly match and invoke target rules in the business rule index library through index identifier (e.g., when the intelligent decision-making end generates a decision requirement of "switching celebrity close-up shots", it automatically matches the corresponding index identifier and invokes the corresponding "priority celebrity close-up" rule). Mapping Matching Channel: Call the rule attribute mapping rule library, and extract the corresponding three-level link affiliation attribute based on the target business rule matched above (e.g., extract the mapped "camera device-camera 1-configuration parameters" based on the star close-up switching rule). Data access channel: Based on the three-level link affiliation attributes extracted by mapping, read the corresponding standardized data from the recording and playback link basic data view (e.g., read the resolution / frame rate under "camera position device - camera position 1 - configuration parameters"). A decision association triggering mechanism is constructed. When the intelligent decision-making terminal triggers a specific business decision requirement, the following chain is completed in sequence: business decision requirement → matching business rules → extracting mapping data attributes → retrieving target data. The target business rules are associated and bound with the corresponding target data to generate a rule-data association group that matches the current decision requirement.

[0019] It should be noted that by classifying the business rules of the recording and playback scenario in three dimensions and assigning unique index identifiers, a structured business rule index library is built. This not only achieves orderly management of business rules, but also greatly improves the calling efficiency of target rules through the index matching mechanism, avoids execution deviations caused by rule chaos, and provides rule guarantee for the system to run accurately according to business needs. The three-level link ownership attribute marking and structured organization of all basic data of the recording and playback link form a standardized basic data view of the recording and playback link. This solves the problems of data dispersion, ambiguous attributes and difficulty in association and retrieval in traditional recording and playback systems, ensures the standardization and traceability of data, and provides high-quality data support for subsequent resource data retrieval during twin simulation and indicator monitoring during real-time optimization. The establishment of the rule attribute mapping library, the three-source input channels, and the decision association triggering mechanism has created a precise correspondence between business rules and data attributes, realizing a closed-loop linkage of "decision requirements - rule matching - data retrieval". The generated rule-data association group directly provides core association basis for the adaptability determination of the twin pre-show scheme optimization module and the abnormal risk prediction of the recording and broadcasting real-time dynamic optimization module, effectively reducing the computational complexity of subsequent modules and ensuring the smooth connection and efficient operation of the entire system process.

[0020] The digital twin pre-rehearsal optimization module constructs a digital twin environment for the live physical system pre-rehearsal, clarifies the business rule requirements for each time node based on the time-series execution baseline, analyzes the rule-data association groups, and determines whether the requirements are suitable. If not, it re-matches the suitable three-level link attribution attributes; integrates the rule-data association groups of all time-series nodes, and outputs the optimized live execution plan. The specific process is as follows: Based on digital twin technology, a digital mirror image corresponding to the physical system of this live broadcast studio (including camera equipment, encoding equipment, material storage units and transmission links) is constructed in a 1:1 ratio to form a digital twin environment for rehearsal; among them, the digital twin environment for rehearsal replicates the three-level link ownership attributes (link component unit - specific device identifier - parameter type) and business rule constraints of the physical system of this live broadcast studio. Obtain the timing execution baseline for this studio live broadcast, which includes: the start time of the official live broadcast, the preset time points of each business process, and the corresponding execution actions; Based on the chronological order, each execution action in the time-series execution baseline is transformed to obtain time-stamped business rule execution requirements; at the same time, the time sequence node corresponding to each time-stamped business rule execution requirement is identified; wherein, the time sequence node corresponds one-to-one with the preset time point in the time-series execution baseline; Start the digital twin environment for the rehearsal, and input each time-stamped business rule execution requirement into the preset decision association triggering mechanism in the digital twin environment for the rehearsal according to the order of the time nodes; The decision-related triggering mechanism calls the business rule index library, rule attribute mapping rule library, and recording and playback link basic data view integrated in the digital twin environment for the pre-drilling. It sequentially completes the complete link of business decision requirements → matching target business rules → extracting the corresponding three-level link attribution attributes → retrieving the target resource data in the current digital image, generating rule-data association groups with time sequence identifiers; among them, the time sequence identifiers correspond one-to-one with the time sequence nodes of the business rule execution requirements with timestamps. A benchmark library for business rules and data association groups is constructed. For each type of business rule, the benchmark library stores the standard three-tier link attribution attribute, the standard resource parameter security range, and the set of alternative three-tier link attribution attributes. The set of alternative three-tier link attribution attributes is preset based on the redundant configuration of the studio equipment. Once each rule-data association group with a time sequence identifier is generated, the target business rule in the rule-data association group is used as the retrieval condition to search the business rule-data association group benchmark library, and the standard three-level link attribution attribute and standard resource parameter security range corresponding to the current target business rule are matched. Then, the three-level link attribution attribute and target resource data in the rule-data association group are compared with the matched standard three-level link attribution attribute and standard resource parameter security range for consistency and adaptability. If the three-level link attribution attribute is completely consistent with the standard attribute, and the target resource data falls within the safe range of the standard resource parameters, the business rule execution requirement of the current time sequence node is deemed to be suitable. The resource consumption pattern data preset by the current target business rule is called. Based on the resource consumption pattern data preset by the current target business rule, the resource data and timestamp of the corresponding three-level link attribution attribute in the digital image are updated. The business rule execution requirement with timestamp corresponding to the next time sequence node in the time sequence execution baseline is read. The above retrieval, matching and comparison operations based on the target business rule as the retrieval condition are continued. If the comparison results do not simultaneously meet the above conditions, it is determined that the business rule execution requirement adaptability of the current time sequence node is unqualified. Using the current target business rule as the search condition, the business rule-data association group benchmark library is searched again to retrieve the corresponding set of candidate three-level link attribution attributes. The candidate three-level link attribution attributes are read in sequence and the resource data corresponding to the candidate three-level link attribution attribute in the digital image is retrieved. At the same time, the standard three-level link attribution attribute and standard resource parameter security range corresponding to the current target business rule in the business rule-data association group benchmark library are matched. For each candidate third-level link attribution attribute read, the formula is used: The matching degree M between the candidate Layer 3 link attribution attributes and the standard Layer 3 link attribution attributes is obtained, where i is the label corresponding to the core performance parameters of the device corresponding to the candidate Layer 3 link attribution attributes and the device corresponding to the standard Layer 3 link attribution attributes, and n is the total number of devices corresponding to the core performance parameters of the device corresponding to the candidate Layer 3 link attribution attributes and the device corresponding to the standard Layer 3 link attribution attributes. For the i-th core performance parameter, let represent the parameter value of the device corresponding to the candidate Layer 3 link affiliation attribute. This refers to the parameter value of the device corresponding to the standard Layer 3 link affiliation attribute of the i-th core performance parameter. The preset weight coefficient corresponding to the i-th core performance parameter (the calculation parameters used in the formula for analyzing the matching degree between the candidate three-level link attribution attribute and the standard three-level link attribution attribute have all been normalized and dimensionless). Key performance parameters include: device encoding rate, link bandwidth capacity, and device response latency; The candidate three-tier link attribution attribute with the highest matching degree is selected from the set of candidate three-tier link attribution attributes. This attribute is then substituted into the benchmark library of the business rule-data association group for matching. The resulting standard three-tier link attribution attribute and the standard resource parameter safety range are compared for compatibility. If the resource data corresponding to the candidate three-tier link attribution attribute falls within the standard resource parameter safety range, it is then used as the best candidate three-tier link attribution attribute to replace the original three-tier link attribution attribute in the rule-data association group. The rule-data association group with time sequence identifier for the current time sequence node is regenerated, and the corresponding resource data and timestamp in the digital image are updated. Then, the business rule execution requirements with timestamp for the next time sequence node are processed. If the resource data corresponding to the candidate third-level link attribution attribute does not fall within the safe range of standard resource parameters, then the candidate third-level link attribution attribute is removed. The candidate third-level link attribution attribute with the highest matching degree is re-selected from the remaining candidate third-level link attribution attributes. The above adaptability comparison operation is repeated until a candidate attribute with qualified adaptability is selected. After completing the rule-data association group update and the update of digital mirror resource data and timestamp for the current time sequence node, the business rule execution requirements with timestamp for the next time sequence node are processed. Once all business rule execution requirements corresponding to all time-series nodes in the time-series execution baseline have completed the above matching, comparison, and processing operations, the final rule-data association group corresponding to all time-series nodes (including those initially generated and deemed qualified, and those regenerated after replacing alternative attributes and deemed qualified) is traversed. The target business rule, the finally determined three-level link attribution attribute, and the updated resource data of each node are extracted in the order of the time-series nodes. The extracted information is integrated into structured data, and the structured data is output as the optimized live broadcast execution scheme.

[0021] It should be noted that the twin pre-playback scheme optimization module, as the "core of pre-optimization" of the system, lays a key foundation for the stable and efficient execution of studio video recording and playback through the deep integration of digital twin technology and adaptive dynamic adjustment mechanism. The digital twin pre-rehearsal environment that replicates the physical system enables "uninterrupted pre-rehearsal" of the entire live broadcast process. It can simulate the execution scenarios of business rules for all time nodes without occupying the resources of the actual recording equipment, and expose potential problems such as configuration conflicts and insufficient resources that are difficult to find in traditional offline debugging in advance. It avoids the risks of stuttering, interruption and abnormal picture caused by improper configuration in the actual live broadcast from the source, and greatly improves the success rate of recording tasks. Based on the demand transformation of the time-series execution baseline and the generation of rule-data association groups with time-series identifiers, a precise binding relationship of "time-rule-data" is established, so that the execution demand of each time-series node has a clear quantitative basis, which solves the pain point of "vague demand and disordered execution" in traditional pre-drills, makes the pre-drill process traceable and reproducible, and provides clear positioning support for subsequent optimization and adjustment. The combination of business rules - data association group benchmark library and matching degree calculation mechanism realizes "precise screening" of adaptation configuration - when the adaptability is not qualified, the optimal alternative three-level link belonging attribute is screened by quantitative comparison of core performance parameters, instead of blindly trying. This ensures that the alternative configuration is highly consistent with the standard attribute, and avoids the problem of "high trial and error cost and low efficiency" in traditional adaptation. At the same time, relying on the redundant configuration of studio equipment pre-equipment selection set, the flexibility and feasibility of adaptation adjustment are ensured. The structured optimization scheme, which integrates the final rules of all time-series nodes and the data association group output, directly provides the "optimal execution blueprint" for the real-time dynamic optimization module of recording and broadcasting. This scheme has completed adaptability verification and configuration optimization in advance, so that no large-scale configuration adjustments are needed during formal recording and broadcasting. Only real-time monitoring and fine-tuning are required, which significantly reduces the computational pressure on the real-time optimization module and ensures the smoothness and efficiency of the overall system operation.

[0022] The real-time dynamic optimization module for recording and playback: Constructs a theoretical standard matrix of time-series and multi-dimensional indicators, and simultaneously constructs an actual monitoring matrix of time-series and multi-dimensional indicators; analyzes the execution adaptation prediction values ​​of business rules for the time-series nodes to be executed, and determines whether there are any execution anomaly risks; if an anomaly is found, a solution is matched, and the adaptation configuration is adjusted until the entire recording and playback monitoring optimization is completed. The specific process is as follows: Based on the optimized live streaming execution scheme, we extract and organize the theoretical values ​​of multi-dimensional core indicators of devices and transmission links corresponding to the three-level link affiliation attributes associated with each time-series node, and construct a time-series-multi-dimensional indicator theoretical standard matrix. Multi-dimensional core indicators include: video bitrate, signal transmission frame rate, device response latency, resource utilization, screen resolution, transmission error rate, storage I / O rate, etc. Among them, the time nodes in the behavioral time-series execution baseline of the time-series multi-dimensional indicator theoretical standard matrix are listed as various multi-dimensional core indicators. The matrix elements are the theoretical standard values ​​of the corresponding time nodes and core indicators, and the preset standard resource parameter safety ranges corresponding to each type of core indicator are marked simultaneously. During the actual recording and broadcasting in the studio, the multi-dimensional core indicator operation data of the executed time-series nodes are collected in real time, and the actual monitoring matrix of time-series-multi-dimensional indicators is constructed simultaneously (the data subject is consistent with the theoretical standard matrix, that is, the actual operation data of the devices and transmission links associated with the executed nodes, and the indicator types correspond one-to-one with the theoretical standard matrix). For the business rules corresponding to the next time node to be executed in the time-series execution baseline, a pre-prediction window is preset (the window range is the preset pre-prediction time interval before the start of the next time node). For each core indicator within the preset pre-judgment window, the data consistency coefficient between the actual monitoring data and the theoretical standard data of the indicator within the pre-judgment window is calculated using the data correlation analysis method, and the data consistency coefficient Sj is obtained. The mean deviation of the index, Dj, is obtained by calculating the arithmetic mean of the absolute differences between all actual monitoring data and corresponding theoretical standard data of the core index within the preset pre-judgment window. The relative fluctuation range Vj of the indicator is obtained by calculating the ratio of the standard deviation of the actual monitored data to the standard deviation of the theoretical standard data within the preset pre-judgment window. After normalizing and removing the dimensions of the data consistency coefficient Sj, the mean deviation of the indicator Dj, and the relative fluctuation of the indicator Vj within the preset prediction window of the current core indicator, the core indicator adaptation score Kj is obtained using the formula: Kj=Sj×b1+Dj×b2+(1-Vj)×b3; where j is the label of the core indicator, j=1,2,……,m; m is the total number of core indicators; b1, b2, b3 are preset weight coefficients; Summarize the fit scores for all core metric types and use the formula: We obtain the execution adaptation prediction value P; where Hj is the preset weight coefficient corresponding to the j-th core indicator; If the pre-judgment value of the adaptation is greater than or equal to the corresponding pre-judgment threshold, it is determined that the business rule corresponding to the next time-series node has the conditions for normal execution. The business rule corresponding to the node is executed according to the optimized live broadcast execution plan. The resource data and timestamp in the actual monitoring matrix of time-series-multi-dimensional indicators and digital image are updated synchronously. The above-mentioned pre-judgment process is continued to be executed on the business rules corresponding to subsequent time-series nodes. If the predicted value for adaptation is less than the adaptation qualification threshold, then the next time-series node is judged to have a risk of execution abnormality. When it is determined that there is a risk of execution anomaly in the business rule corresponding to the next time sequence node, the multi-dimensional core comparison parameters in the preset preceding prediction window corresponding to the business rule are extracted, including: data consistency coefficient of each type of core indicator, average indicator deviation, relative fluctuation of indicator, core indicator adaptation score, execution adaptation prediction value, etc. The extracted multi-dimensional core comparison parameters are organized to construct a set of abnormal features corresponding to the business rules that currently have the risk of execution anomalies; The abnormal feature set is subjected to three-dimensional feature dimensionality reduction processing using principal component analysis to obtain a low-dimensional abnormal feature vector. Construct an anomaly feature-solution mapping library; for each type of low-dimensional anomaly feature vector corresponding to each business rule, the library pre-sets and stores corresponding targeted solutions (solutions include: three-level link affiliation attribute switching, resource configuration parameter adjustment, redundant equipment activation, transmission protocol optimization, etc.). Substitute the low-dimensional abnormal feature vector corresponding to the business rule that currently has the risk of execution anomaly into the abnormal feature-solution mapping library. Calculate the degree of fit between the low-dimensional abnormal feature vector and various preset abnormal feature vectors in the library using the feature vector cosine similarity matching algorithm. Select the target solution with the highest degree of fit and send the target solution to the operation and maintenance management terminal. Based on the target solution, the operators of the operation and maintenance management terminal simulate and verify the adjustment effect in the digital image, adjust the three-level link attribution attribute or resource configuration parameters corresponding to the business rule, regenerate the rule-data association group of the business rule, correct the core indicator theoretical value of the business rule corresponding to the next time-series node in the time-series-multi-dimensional indicator theoretical standard matrix, repeat the above prediction process until the execution adaptation prediction value is greater than or equal to the adaptation qualification threshold, and then execute the business rule corresponding to the next time-series node. Repeat the above prediction, execution, exception handling and solution update process until all business rules corresponding to all time nodes in the time sequence execution baseline are executed, and complete the real-time monitoring and dynamic optimization of the entire studio's formal recording and broadcasting.

[0023] It should be noted that the real-time dynamic optimization module for recording and broadcasting, as the "real-time guarantee core" of the system, has achieved an upgrade from "passive remediation" to "proactive prevention and control" in studio recording and broadcasting through a full-process mechanism of "quantitative prediction - precise measures - closed-loop optimization". The dual-matrix construction achieves "quantitative benchmarking," overcoming the pain point of "fuzziness" in traditional recording and broadcasting monitoring: the time-series-multi-dimensional indicator theoretical standard matrix clarifies the quantitative standards and safety ranges of core indicators at each time-series node, while the time-series-multi-dimensional indicator actual monitoring matrix synchronously collects real-time operating data, forming a precise benchmarking relationship between the two. The multi-dimensional core indicators comprehensively cover key dimensions such as video quality (bitrate, resolution), equipment performance (resource utilization, response latency), and transmission stability (bit error rate, frame rate), ensuring comprehensive monitoring without blind spots. This provides comprehensive and quantitative data support for risk prediction, avoiding risk misjudgment or omissions caused by monitoring a single indicator. The pre-judgment window and multi-parameter fusion calculation realize "early warning": by extracting key parameters such as data consistency coefficient, average index deviation, and relative fluctuation amplitude within the prediction window, and after normalization and weight fusion, the execution adaptation prediction value is obtained. This can identify potential abnormal risks in advance before the execution of the next-level node, completely changing the traditional system's "post-event discovery and passive remedy" mode, reserving sufficient time for anomaly handling, and greatly reducing the probability of serious problems such as recording interruption and abnormal picture. The anomaly handling mechanism achieves "precise policy implementation" while balancing efficiency and security: by constructing an anomaly feature set, extracting core features through 3D dimensionality reduction, and combining cosine similarity matching of the anomaly feature-solution mapping library, it can quickly locate the most suitable solution, avoiding the problems of "blind adjustment and high trial-and-error costs" in traditional anomaly handling; at the same time, maintenance personnel can simulate and verify the adjustment effect in a digital image without directly operating the actual recording equipment, which not only ensures the feasibility of the adjustment plan, but also avoids the secondary risks that may be caused by on-site adjustments, achieving a dual guarantee of "safe adjustment + efficient policy implementation"; Closed-loop optimization and full-process iteration ensure "end-to-end stability": The module continuously optimizes each time-series node through a closed-loop process of "prediction-execution-anomaly handling-correction-re-prediction." This not only resolves anomalies at the current node but also provides a more quantitative basis for subsequent nodes' execution that better reflects actual operating conditions by revising the core indicator values ​​of the theoretical standard matrix, forming a dynamically iterative optimization system. This process continues until all time-series nodes are completed, achieving continuous monitoring and dynamic adjustment of the entire recording process and ensuring the stability and high quality of the final recording output. The module connects to form a complete closed loop of "pre-optimization + real-time support": This module constructs a theoretical standard matrix based on the optimized execution scheme output by the twin pre-performance, so that real-time optimization has a precise initial basis; at the same time, the configuration and indicator data corrected during the real-time optimization process can also provide a reference for the pre-performance optimization of subsequent recording and broadcasting tasks, forming cross-module experience accumulation and iterative upgrades, and continuously improving the overall operating efficiency and optimization effect of the system.

[0024] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A studio video recording and playback system, characterized in that, include: Rule data system construction module: acquire business rules for recording and playback scenarios, build a business rule index library; collect full data of the recording and playback chain, and build a basic data view of the recording and playback chain; Build a rule attribute mapping library, establish a three-source input channel and decision association mechanism, and generate rule-data association groups that match decision requirements; Twin rehearsal optimization module: Construct a digital twin environment for live physical system rehearsal, clarify the business rule requirements of each time node based on the time-series execution baseline, analyze the rule-data association group, and determine whether the requirement adaptability is qualified; if not qualified, rematch the adaptable three-level link attribution attribute; integrate the rule-data association group of all time-series nodes, and output the optimized live execution plan; Recording and playback real-time dynamic optimization module: Constructs a theoretical standard matrix of time-series and multi-dimensional indicators, and simultaneously constructs an actual monitoring matrix of time-series and multi-dimensional indicators; analyzes the execution adaptation prediction value of the business rules of the time-series nodes to be executed, and judges whether there is any risk of execution anomalies; if there is an anomaly, a solution is matched, the adaptation configuration is adjusted, until the full-process recording and playback monitoring optimization is completed.

2. The studio video recording and playback system according to claim 1, characterized in that: The construction process of the business rule index library, the basic data view of the recording and playback link, and the rule attribute mapping library is as follows: Obtain the business rules corresponding to the current recording and playback scenario, classify and organize them according to three dimensions: scenario allocation type, preset rule priority, and constraint object, build a business rule index library, and assign a unique index identifier to each type of business rule; Real-time collection of all basic data of each component of the recording and playback link, marking each data item with the corresponding three-level link affiliation attribute, and classifying them according to the attribute and collection timestamp to form a structured basic data view of the recording and playback link; Construct a rule attribute mapping library, which pre-sets and stores the three-level link attribution attributes corresponding to each type of business rule in the business rule index library.

3. The studio video recording and playback system according to claim 2, characterized in that: The specific process of establishing a three-source input channel and decision association mechanism to generate rule-data association groups that match decision requirements is as follows: The business rule index library, rule attribute mapping library, and basic data view of recording and playback link are used as the three sources of input for the intelligent decision-making terminal, and the rule calling channel, mapping matching channel and data calling channel are constructed simultaneously. The rule invocation channel invokes the target rule through the decision requirement-index identifier matching logic. The mapping matching channel extracts the three-level link attribution attribute corresponding to the target rule. The data invocation channel reads the standardized data corresponding to this attribute. A decision association triggering mechanism is constructed, which follows the business decision requirement → matching business rules → extracting mapping data attributes → invoking the target data link, binding the target business rule with the corresponding data, and generating a rule-data association group.

4. A studio video recording and playback system according to claim 3, characterized in that: The process of constructing a digital twin environment for live streaming physical system rehearsals, clarifying business rule requirements for each time node based on the time-series execution baseline, and analyzing rule-data association groups is as follows: A digital mirror image corresponding to the studio's live broadcast physical system is constructed based on digital twin technology to form a digital twin environment for rehearsals; Obtain the execution baseline based on the time sequence, transform the execution actions into business rule execution requirements with timestamps according to the time sequence, and clarify the corresponding time sequence nodes; Start the pre-rehearsal environment and input the requirements into the built-in decision association triggering mechanism in sequence. This mechanism calls the integrated business rule index library, rule attribute mapping library and the basic data view of the recording and playback link in the digital image. Through matching rules → extracting attributes → calling data links, a rule-data association group with time sequence identifier is generated.

5. A studio video recording and playback system according to claim 4, characterized in that: The specific process for determining whether the business rule requirements are compatible is as follows: Construct a benchmark library for business rules and data association groups. The library stores the standard three-tier link attribution attributes, standard resource parameter security ranges, and alternative three-tier link attribution attribute sets corresponding to each type of business rule. After the rule-data association group with time sequence identifier is generated, the benchmark library is searched with the target business rule to match the corresponding standard attributes and security intervals. The three-level link affiliation attributes and target resource data in the association group are compared with them respectively. If the attributes are completely consistent and the data is within the safe range, the compatibility is deemed satisfactory, the corresponding resource data and timestamp of the digital image are updated, and the requirements of the next time-series node are processed; otherwise, the compatibility is deemed unsatisfactory.

6. A studio video recording and playback system according to claim 5, characterized in that: If the match is unsuccessful, a new matching of the three-tier link attribution attributes will be performed. When the compatibility is not up to standard, the business rule-data association group benchmark library is retrieved based on the current target business rule, the corresponding set of alternative three-level link attribution attributes is retrieved, each alternative three-level link attribution attribute is read in turn and the corresponding resource data in the digital image is obtained, and the standard three-level link attribution attribute and standard resource parameter security range corresponding to the business rule in the business rule-data association group benchmark library are matched at the same time. By combining the parameter values ​​of core performance parameters with preset weight coefficients, the matching degree between the attribution attributes of each candidate third-level link and the attribution attributes of the standard third-level link is calculated. The candidate third-level link attribution attributes with the highest matching degree are selected, and their corresponding resource data is compared with the standard resource parameter security range in the benchmark library of business rules-data association group. If the data is within the safe range, replace the original three-level link attribution attribute in the rule-data association group with the alternative three-level link attribution attribute, regenerate the rule-data association group with time sequence identifier, update the digital mirror resource data and timestamp, and continue to process the next time sequence node requirement; If the data is not within the safe range, remove the candidate third-level link attribution attribute, and repeat the screening and comparison until a suitable candidate third-level link attribution attribute is found and the update is completed.

7. A studio video recording and playback system according to claim 6, characterized in that: The specific process of integrating the rules and data association groups of all time-series nodes and outputting the optimized live streaming execution plan is as follows: Once all business rule execution requirements corresponding to all time-series nodes in the time-series execution baseline have completed the above matching, comparison, and processing operations, the final rule-data association group corresponding to all time-series nodes is traversed. The target business rule, the finally determined three-level link attribution attribute, and the updated resource data of each node are extracted in the order of the time-series nodes. The extracted information is integrated into structured data, and the structured data is output as the optimized live broadcast execution scheme.

8. A studio video recording and playback system according to claim 7, characterized in that: The specific process for analyzing the execution adaptation prediction values ​​of the business rules to be executed at the time sequence nodes and determining whether there is a risk of execution anomalies is as follows: Based on the optimized live streaming execution scheme, we extract the theoretical values ​​of multi-dimensional core indicators of devices and transmission links corresponding to the three-level link affiliation attributes associated with each time-series node, and construct a time-series-multi-dimensional indicator theoretical standard matrix. During the actual recording and broadcasting, multi-dimensional core indicator operation data of the executed nodes are collected in real time, and a time-series-multi-dimensional indicator actual monitoring matrix is ​​constructed simultaneously. Pre-set a preceding prediction window for the business rules of the next time sequence node, extract the consistency coefficient between actual and theoretical data, the average deviation of indicators, and the relative fluctuation range of each type of core indicator in the window, and after normalization, calculate the core indicator adaptation score in combination with the preset weight coefficient, and then summarize the score and combine it with the corresponding preset weight coefficient to obtain the execution adaptation prediction value. If the predicted value is not lower than the preset suitable qualification threshold, it is determined that the conditions for normal execution are met, and the plan is executed and the relevant matrices and data are updated; if it is lower, it is determined that there is a risk of execution abnormality.

9. A studio video recording and playback system according to claim 8, characterized in that: If an anomaly is detected, a solution will be matched, and the adaptation configuration will be adjusted until the entire recording and playback monitoring optimization process is completed. The specific process is as follows: When an execution anomaly risk is determined, the multi-dimensional core comparison parameters in the preceding prediction window corresponding to the business rule are extracted, an anomaly feature set is constructed, and a low-dimensional anomaly feature vector is obtained by principal component analysis and three-dimensional dimensionality reduction. Substitute it into the anomaly feature-solution mapping library, filter the target solution with the highest matching degree through feature vector cosine similarity matching, and send it to the operation and maintenance management terminal; Operators simulate and verify the target solution in the digital mirror, adjust the three-level link attribution attribute or resource configuration parameters corresponding to the business rules, regenerate the rule-data association group and correct the theoretical values ​​of the core indicators related to the time series-multi-dimensional indicator theoretical standard matrix, repeat the prediction process until the prediction value meets the standard, and then execute the business rules of the next time series node. Repeat the above process until all time-series node business rules have been executed, thus completing the optimization of the entire recording and broadcasting monitoring process.

Images