CN120218561A - A data monitoring method and system based on image recognition - Google Patents
A data monitoring method and system based on image recognition Download PDFInfo
- Publication number
- CN120218561A CN120218561A CN202510652444.1A CN202510652444A CN120218561A CN 120218561 A CN120218561 A CN 120218561A CN 202510652444 A CN202510652444 A CN 202510652444A CN 120218561 A CN120218561 A CN 120218561A
- Authority
- CN
- China
- Prior art keywords
- translation
- adjustment factor
- behavior
- index
- priority value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063114—Status monitoring or status determination for a person or group
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Psychiatry (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
本发明适用于图像识别与翻译任务调度技术领域，提供了一种基于图像识别的数据监控方法及系统，所述方法包括：在确定目标译员存在特定设备切换行为时，获取目标译员的原始调度优先级值、历史翻译记录以及当前图像片段，并确定当前翻译任务的类型。本发明通过引入设备适应指数与翻译投入指数两个行为调整因子，构建了一种面向设备切换场景的译员调度优先级动态修正机制。相比传统仅依赖静态评分或单次状态评估的方式，本发明能够同时感知译员在跨设备任务中的长期适应趋势与当前任务的即时切入效率，提升调度判断的精准性与鲁棒性。
The present invention is applicable to the technical field of image recognition and translation task scheduling, and provides a data monitoring method and system based on image recognition, the method comprising: when it is determined that the target translator has a specific device switching behavior, obtaining the original scheduling priority value, historical translation records and current image fragment of the target translator, and determining the type of the current translation task. The present invention constructs a dynamic correction mechanism for translator scheduling priority for device switching scenarios by introducing two behavior adjustment factors, namely the device adaptation index and the translation input index. Compared with the traditional method that only relies on static scoring or single state evaluation, the present invention can simultaneously perceive the long-term adaptation trend of the translator in cross-device tasks and the immediate entry efficiency of the current task, thereby improving the accuracy and robustness of scheduling judgments.
Description
Technical Field
The invention belongs to the technical field of image recognition and translation task scheduling, and particularly relates to a data monitoring method and system based on image recognition.
Background
In existing translation task allocation systems, a fixed scheduling policy or static scoring mechanism is typically employed to task the translator. The system is used for carrying out priority ordering on the basis of the language direction, the historical evaluation score, the task completion rate and other dimensions of the translator, so that automatic matching of tasks is realized. However, the existing method generally lacks the perception capability of the current actual state of the translator, and cannot judge whether the translator is in a working state suitable for accessing the task immediately, especially in a multi-terminal mixed environment, the system cannot effectively identify whether the translator has completed adapting to the current equipment, so that state deviation between scheduling and execution may occur.
Meanwhile, some technologies attempt to introduce a real-time video monitoring or behavior recognition module to acquire online state or simple behavior change of a translator, but the methods often stay in rough recognition of whether on duty or active, and cannot establish a dynamic evaluation model from deep features such as behavior evolution trend, task cut-in efficiency and the like, and also do not establish a systematic behavior analysis mechanism for key interference factors common in remote or multi-terminal collaborative translation tasks such as device switching. Therefore, the prior art has obvious problems of perception blind areas and scheduling delay when dealing with complex working scenes such as frequent cross-platform work of translators, uncertain access states and the like.
Disclosure of Invention
The invention aims to provide a data monitoring method and system based on image recognition, and aims to solve the problems in the background technology.
The invention is realized in such a way that a data monitoring method based on image recognition comprises the following steps:
when the fact that the target translator has specific equipment switching behaviors is determined, an original scheduling priority value, a historical translation record and a current image fragment of the target translator are obtained, and the type of a current translation task is determined;
based on the current image fragment, performing image recognition processing on the target translator to generate a current behavior label of the target translator;
According to the type of the current translation task, the current behavior label and the switching behavior of the specific equipment, a plurality of associated translation fragments are screened out from the historical translation record;
Based on the historical translation record, extracting the equipment adaptation index of the target translator in each associated translation segment, and calculating and generating a first adjustment factor according to the time sequence change trend of the equipment adaptation index;
performing intelligent evaluation on each associated translation segment, generating a corresponding translation input index, acquiring a preset reference index, and calculating to obtain a second adjustment factor based on the difference between each translation input index and the preset reference index;
And comprehensively correcting the original scheduling priority value based on the first adjustment factor and the second adjustment factor.
As a further limitation of the technical solution of the embodiment of the present invention, based on the history translation record, the step of extracting the device adaptation index of the target translator in each associated translation segment, and calculating and generating the first adjustment factor according to the time sequence variation trend of the device adaptation index includes:
analyzing the historical translation record, and extracting image behavior fragments of a target translator in a preset period after completing the switching behavior of specific equipment in each associated translation fragment;
Analyzing the image behavior segments based on an image recognition technology, extracting behavior parameters of face orientation, lip movement rhythm, gesture stability and eye movement track, calculating convergence degree of the behavior parameters in a preset period, generating corresponding behavior stability scores, and weighting and fusing the stability scores to generate equipment adaptation indexes;
and constructing a time sequence based on the device adaptation indexes corresponding to the associated translation fragments, and taking the average slope of the time sequence as a first adjustment factor.
As further defined by the technical solution of the embodiment of the present invention, the steps of intelligently evaluating each associated translation segment, generating a corresponding translation input index, obtaining a preset reference index, and calculating to obtain a second adjustment factor based on a difference between each translation input index and the preset reference index include:
Analyzing each associated translation segment, performing behavior monitoring on a target translator at the translation task starting stage, identifying the behavior change process of the target translator from task starting to stable translation state entering, measuring the duration time required by the target translator to reach the translation concentration state, and generating a corresponding translation input index based on the duration time;
acquiring a preset reference index, wherein the preset reference index refers to average input duration counted in a plurality of high-quality translation behavior fragments;
and calculating the deviation amplitude value of each translation input index compared with the preset reference index, and taking the average value of all the deviation amplitude values as a second adjustment factor.
As a further limitation of the technical solution of the embodiment of the present invention, the step of comprehensively correcting the original scheduling priority value based on the first adjustment factor and the second adjustment factor includes:
a preset priority value correction formula is called, and the original scheduling priority value is subjected to weighted correction by combining the first adjustment factor and the second adjustment factor to obtain a corrected scheduling priority value;
and applying the corrected scheduling priority value to a task allocation flow of the target translator, and optimizing the scheduling sequence or task matching decision of the target translator.
As a further limitation of the technical solution of the embodiment of the present invention, the priority value correction formula is: wherein Refers to the revised scheduling priority value,Referring to the original scheduling priority value,Refers to the first adjustment factor, i.e., the average slope of the time series,Refers to the adjustment weight of the first adjustment factor,Refers to the total number of associated translated segments,Refers to the firstThe translation input index corresponding to each associated translation segment,Refers to the preset reference index of the index,Refers to the second adjustment factor, i.e. the average of all deviation magnitudes,Refers to the adjustment weight of the second adjustment factor, andAndAre all greater than 0.
As a further limitation of the technical scheme of the embodiment of the invention, the data monitoring system based on image recognition comprises a data acquisition module, a label setting module, a data screening module, a first adjustment factor determining module, a second adjustment factor determining module and a priority value correcting module, wherein:
The data acquisition module is used for acquiring an original scheduling priority value, a historical translation record and a current image fragment of the target translator when determining that the target translator has specific equipment switching behaviors, and determining the type of a current translation task;
The label setting module is used for carrying out image recognition processing on the target translator based on the current image fragment to generate a current behavior label of the target translator;
The data screening module is used for screening a plurality of associated translation fragments from the historical translation record according to the type of the current translation task, the current behavior label and the switching behavior of the specific equipment;
The first adjustment factor determining module is used for extracting the equipment adaptation index of the target translator in each associated translation segment based on the historical translation record, and calculating and generating a first adjustment factor according to the time sequence change trend of the equipment adaptation index;
The second adjustment factor determining module is used for intelligently evaluating each associated translation segment, generating a corresponding translation input index, acquiring a preset reference index, and calculating to obtain a second adjustment factor based on the difference between each translation input index and the preset reference index;
And the priority value correction module is used for comprehensively correcting the original scheduling priority value based on the first adjustment factor and the second adjustment factor.
As a further limitation of the technical solution of the embodiment of the present invention, the first adjustment factor determining module specifically includes:
The image segment extraction unit is used for analyzing the history translation record and extracting image behavior segments in a preset period after the target translator completes the switching behavior of the specific equipment in each associated translation segment;
The device adaptation index generation unit is used for analyzing the image behavior segments based on the image recognition technology, extracting behavior parameters of face orientation, lip movement rhythm, gesture stability and eye movement track, calculating the convergence degree of each behavior parameter in a preset period, generating corresponding behavior stability scores, and weighting and fusing each stability score to generate a device adaptation index;
And the average slope calculation unit is used for constructing a time sequence based on the device adaptation indexes corresponding to the associated translation fragments, and taking the average slope of the time sequence as a first adjustment factor.
As a further limitation of the technical solution of the embodiment of the present invention, the second adjustment factor determining module specifically includes:
The translation input index generation unit is used for analyzing each associated translation fragment, performing behavior monitoring on a target translator at the translation task starting stage, identifying the behavior change process of the target translator from task starting to stable translation state, measuring the duration time required by the target translator to reach the translation concentration state, and generating a corresponding translation input index based on the duration time;
the preset reference index acquisition unit is used for acquiring a preset reference index, wherein the preset reference index refers to average input time counted in a plurality of high-quality translation behavior fragments;
and the deviation amplitude calculation unit is used for calculating the deviation amplitude of each translation input index compared with the preset reference index, and taking the average value of all the deviation amplitudes as a second adjustment factor.
As a further limitation of the technical solution of the embodiment of the present invention, the priority value correction module specifically includes:
the priority value correction unit is used for calling a preset priority value correction formula, and combining the first adjustment factor and the second adjustment factor to carry out weighted correction on the original scheduling priority value so as to obtain a corrected scheduling priority value;
And the corrected priority value application unit is used for applying the corrected dispatching priority value to the task allocation flow of the target translator and optimizing the dispatching sequence or task matching decision of the target translator.
As a further limitation of the technical solution of the embodiment of the present invention, the priority value correction formula is: wherein Refers to the revised scheduling priority value,Referring to the original scheduling priority value,Refers to the first adjustment factor, i.e., the average slope of the time series,Refers to the adjustment weight of the first adjustment factor,Refers to the total number of associated translated segments,Refers to the firstThe translation input index corresponding to each associated translation segment,Refers to the preset reference index of the index,Refers to the second adjustment factor, i.e. the average of all deviation magnitudes,Refers to the adjustment weight of the second adjustment factor, andAndAre all greater than 0.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, by introducing two behavior adjustment factors of the equipment adaptation index and the translation input index, a translator scheduling priority dynamic correction mechanism facing the equipment switching scene is constructed. Compared with the traditional mode of only relying on static scoring or single state evaluation, the method and the device can simultaneously sense the long-term adaptation trend of the translator in the cross-equipment task and the instant cut-in efficiency of the current task, and improve the accuracy and the robustness of scheduling judgment. The mechanism is suitable for the translation task distribution in the multi-terminal collaborative environment, can effectively avoid unstable behavior or response delay caused by equipment switching, improves the intelligent level of system scheduling and the reasonability of translator matching, and has obvious practical value and popularization prospect.
Drawings
FIG. 1 is a flow chart of a method provided by an embodiment of the present invention;
FIG. 2 is a flowchart of a method for generating a first adjustment factor according to an embodiment of the present invention;
FIG. 3 is a flowchart of a method for generating a second adjustment factor according to an embodiment of the present invention;
FIG. 4 is a flowchart for correcting an original scheduling priority value in a method according to an embodiment of the present invention;
FIG. 5 is an application architecture diagram of a system provided by an embodiment of the present invention;
FIG. 6 is a block diagram illustrating a first adjustment factor determination module in a system according to an embodiment of the present invention;
FIG. 7 is a block diagram illustrating a second adjustment factor determination module in a system according to an embodiment of the present invention;
fig. 8 is a block diagram of a priority value correction module in the system according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Fig. 1 shows a flowchart of a method provided by an embodiment of the present invention.
Specifically, the method for monitoring data based on image recognition specifically comprises the following steps:
step S100, when the fact that the target translator has specific equipment switching behaviors is determined, the original scheduling priority value, the historical translation record and the current image fragment of the target translator are obtained, and the type of the current translation task is determined.
In the embodiment of the invention, the method is suitable for a scheduling scene which comprises a plurality of translators and is dynamically allocated for the tasks. In general, the system is applied to a remote or multi-platform translation task environment, and a plurality of translators are in a waiting state and wait for the system to distribute new translation tasks. Each translator may be located in a different terminal device (such as a desktop, a notebook, a tablet, a mobile terminal, etc.), and the system needs to dynamically evaluate its fitness according to the actual state and the historical performance of the translator, and accordingly optimize the translator scheduling priority, so as to implement the person post matching optimization of the translation task.
The specific device switching behavior refers to the behavior that a target translator switches from one translation terminal device to another device in a short time, for example, from a mobile device to a desktop terminal, or from an offline co-transmission system to a remote video platform, etc. Because the change of the equipment environment generally affects the operation fluency, the behavior rhythm and the task adaptability of the translator, the behavior recognition mechanism is introduced in the scheduling decision, and the task can be effectively prevented from being distributed to the translator which has not completed the adaption or has unstable operation, thereby improving the translation quality and the reliability of the system scheduling.
The original scheduling priority value is a reference ordering basis preset by the system for a translator in a state of not distributing tasks currently, and can be generally obtained by an existing translation platform management system, an intelligent scheduling system or a task matching module in a statistical scoring mode. For example, the method can perform preliminary evaluation based on factors such as language matching degree, historical completion rate, average response time, feedback quality score and the like of the translator, generate an initial priority score, and have good practical availability and engineering realization foundation in the existing remote translation service platform.
The historical translation record is a data set formed by the target translator in the past task execution process, and generally comprises information such as historical task numbers of the translator, corresponding task types, participation time, equipment environments, behavior tracks (such as image frame sequences), input performance data (such as time taken from the start of a task to the concentration state), equipment adaptation related performance (such as operation stability after equipment switching) and the like. The basic data are taken as the basis of behavior modeling and trend extraction, and are the original sources of the adaptation index and the translation input index of the subsequent extraction equipment.
The current image segment is derived from an image acquisition device of the current terminal of the target translator and generally comprises a video frame sequence acquired by a camera module. The system periodically acquires the image data of the translator during the period of waiting for task allocation, and intercepts a continuous image frame sequence of a specific period (for example, 30-60 seconds after switching) after recognizing that the device switching behavior exists, and the continuous image frame sequence is used as an input segment for subsequent image recognition processing and behavior feature extraction.
The type of the current translation task can be judged by the dispatching platform based on the task source information, and generally comprises classification information such as task subject, language direction, service type (such as law, medical treatment, technology, conference and the like), task duration or emergency degree and the like, wherein the metadata is generally provided by a task initiating terminal or a user interface when the task is generated or accessed into a system, belongs to the category of task description information existing in the existing platform, and can be used as one of context conditions in dispatching decision.
Further, the data monitoring method based on image recognition further comprises the following steps:
step S200, performing image recognition processing on the target translator based on the current image fragment to generate a current behavior label of the target translator.
In the embodiment of the invention, the core objective of the step S200 is to perform image recognition analysis on the target translator based on the current image fragment and judge the current working state of the target translator according to the image recognition analysis, so as to match a behavior label with semantic description for the target translator for subsequent label comparison and behavior trend evaluation of the historical translation fragment.
The step mainly relies on the existing image recognition and gesture recognition technology. The system acquires the current image fragment of the target translator through the camera, namely, video frame data in a period of time window after the switching action of the specific equipment is completed, and extracts key point parameters including face orientation, eye movement direction, head rotation angle, mouth opening and closing condition, shoulder and hand position change and the like by utilizing an attitude estimation algorithm (such as OpenPose, mediaPipe and the like). And further extracting characteristic indexes such as action persistence, frequency, amplitude and the like from the image frame sequence.
In combination with these features, the system may determine whether the target translator is currently in a steady rest, frequent motion, turn-around conversation, eating, long-term low head, etc., state, and match the corresponding behavioral label accordingly. Behavior tags may include, but are not limited to, the following categories:
"sitting in mind", "non-interactive state", "eating", "walking", "inattention", "communicating with others", "device adjustment", "hand frequent movements", "away from the device" etc.
The generation of these labels depends on a classification rule or a lightweight behavior classification model obtained by training, and can be constructed based on the existing image samples by a supervised learning mode. The tag is not used for reasoning the translation quality of the translator, but is only used as semantic expression of behavior characteristics, and is used as a matching condition in the subsequent historical behavior screening, so that the historical fragments are guaranteed to be highly similar to the current translator state, and the generation of a device adaptation index and a translation input index with more reference values is facilitated.
Further, the data monitoring method based on image recognition further comprises the following steps:
Step S300, a plurality of associated translation fragments are screened from the historical translation records according to the type of the current translation task, the current behavior label and the switching behavior of the specific equipment.
In the embodiment of the present invention, the purpose of step S300 is to screen a plurality of relevant translation segments matching the current situation from the historical translation records of the target translator, as the data base of the adaptation index and the translation input index of the subsequent computing device, so as to improve the pertinence and the effectiveness of the behavior trend evaluation.
The specific screening process includes the following aspects:
Firstly, the system determines the semantic category to which the task belongs, such as legal category, medical category, conference category, technical category and the like, according to the type of the current translation task. The task type is provided by task metadata as one of the preliminary filtering conditions for filtering out historical segments in the historical translation record that are significantly inconsistent with the current task type.
Next, the system calls the current behavior tag identified in step S200 as a second-layer screening condition. The system extracts all translation fragments with the same behavior label in the initial stage of task access from the historical translation record, for example, the current label is 'sitting still focus', and only the fragments with the same "sitting still focus" label in the initial stage of the task in the history are reserved, so that the behavior states are ensured to be comparable.
Again, the system further defines that the history translation record only contains translation segments with device switching types consistent with the current state in combination with the specific device switching behavior currently identified, such as "switch from mobile terminal to desktop terminal" or "switch from offline terminal to remote platform". The device switching behavior can be matched through the task access log or the device change record in the history behavior record, so that the screening result is ensured to have structural consistency in the aspect of device adaptability.
Finally, on the basis of meeting the three conditions of task type matching, behavior label consistency and equipment switching type consistency, the system selects a plurality of translation fragments meeting the conditions from the history translation record as associated translation fragments, sorts and numbers the translation fragments according to time sequence or task quality level, and provides input data for calculating equipment adaptation index and translation input index in the subsequent step.
Through the screening mechanism, the system can ensure that the extracted historical behavior sample is highly similar to the actual state of the current translator, so that the accuracy of behavior trend modeling and the reference value of priority correction are improved.
Further, the data monitoring method based on image recognition further comprises the following steps:
step S400, extracting the equipment adaptation index of the target translator in each associated translation segment based on the historical translation record, and calculating and generating a first adjustment factor according to the time sequence change trend of the equipment adaptation index.
Specifically, fig. 2 shows a flowchart of generating the first adjustment factor.
The method for generating the first adjustment factor by calculating according to the time sequence change trend of the equipment adaptation index specifically comprises the following steps of:
Step S401, analyzing the history translation record, and extracting image behavior fragments in a preset period after the target translator completes the switching behavior of the specific equipment in each associated translation fragment;
Step S402, analyzing the image behavior segments based on an image recognition technology, extracting behavior parameters of face orientation, lip movement rhythm, gesture stability and eye movement track, calculating convergence degree of the behavior parameters in a preset period, generating corresponding behavior stability scores, and weighting and fusing the stability scores to generate equipment adaptation indexes;
Step S403, a time sequence is constructed based on the device adaptation indexes corresponding to the associated translation segments, and the average slope of the time sequence is used as a first adjustment factor.
In the embodiment of the present invention, step S402 parses the image behavior segment by using an image recognition technology, and extracts a plurality of behavior parameters of the target translator within a preset period after the specific device is switched. These behavioral parameters, including facial orientation, lip movement cadence, gesture stability, and eye movement trajectories, can be extracted by existing image pose estimation and feature tracking algorithms, such as recognition based on OpenPose, mediaPipe or other deep learning based skeletal key point detection models. The system identifies facial deflection angle, mouth opening and closing frequency, amplitude fluctuation of upper body gesture and eyeball line of sight change condition through continuous image analysis of head, eyes, mouth, shoulders and other key areas in the video frame.
In a set time period, the system constructs a parameter sequence for each behavior parameter, and calculates stability indexes of the sequence, such as variation amplitude, fluctuation frequency, maximum and minimum value difference and the like, so as to judge whether the sequence shows a variation trend from fluctuation to stability. The system judges the convergence degree of each type of parameter, namely whether the behavior of the equipment after switching tends to be stable in a short time. After each parameter is respectively endowed with a stability score, the scores of the multiple items are weighted and fused according to a preset weight, so that a device adaptation index comprehensively representing the adaptation capability of the translator under the current device switching condition is generated.
In step S403, the system arranges the device adaptation indexes generated in the plurality of associated translation segments in time order, and constructs a time sequence of the device adaptation indexes. The time sequence can reflect whether the adaptive performance of the translator has obvious trend after the equipment is switched in different historical tasks. And the system carries out linear fitting on the time sequence, and obtains the average slope of the fitting straight line as a first adjustment factor.
The core reason for using the average slope as the first adjustment factor is that the slope may quantify the trend of the change in the adaptability of the translator device. If the slope is positive, the performance of the translator under the condition of equipment switching is improved gradually, the translator has an adaptability lifting trend, if the slope is negative, the adaptability is reduced, and if the slope is close to zero, the translator can be regarded as the equipment adaptability is stable. In this way, the system can dynamically perceive the adaptive evolution trend of the target translator in the switching tasks of a plurality of devices, thereby providing a quantifiable and trend-oriented adjustment basis for the correction of the priority value.
Further, the data monitoring method based on image recognition further comprises the following steps:
and S500, performing intelligent evaluation on each associated translation segment, generating a corresponding translation input index, acquiring a preset reference index, and calculating to obtain a second adjustment factor based on the difference between each translation input index and the preset reference index.
Specifically, fig. 3 shows a flowchart for generating the second adjustment factor.
Wherein, each associated translation segment is intelligently evaluated, a corresponding translation input index is generated, a preset reference index is obtained, and based on the difference between each translation input index and a preset reference index, calculating a second adjustment factor specifically comprises the following steps:
Step S501, analyzing each associated translation segment, performing behavior monitoring on a target translator at the translation task starting stage, identifying the behavior change process of the target translator from task starting to stable translation state entering, measuring the duration time required by the target translator to reach the translation concentration state, and generating a corresponding translation input index based on the duration time;
Step S502, obtaining a preset reference index, wherein the preset reference index refers to average input duration counted in a plurality of high-quality translation behavior fragments;
In step S503, the deviation amplitude of each translation input index compared with the preset reference index is calculated, and the average value of all the deviation amplitudes is used as the second adjustment factor.
In the embodiment of the present invention, the system identifies the behavior change process undergone by the target translator from task initiation to entering the stable translation state by performing continuous behavior monitoring on the image behavior sequence of the associated translation segment in step S501. Specifically, the system analyzes continuous image frames after the translation task starts based on an image recognition technology, and extracts key action features of regions such as head, face, eyes, mouth, shoulders and the like, including whether face orientation is stable, whether eye concentration is focused on a screen, whether the mouth is regularly opened and closed, whether the upper body is static and the like.
During behavior monitoring, the system recognizes the onset of gradual transition of translator behavior from unstable to stable by setting a set of behavior stability thresholds, e.g., face offset angles continuously below a set range, mouth cadence maintained at a fixed frequency, line of sight continuously concentrated in screen areas, etc. When these behavior parameters meet the stability requirement for the first time in the image sequence, the system defines this point in time as the decision node for entering the stable translation state, and calculates the length of time between the start of the task and this node, i.e. the duration of time required for the translator to complete the state cut.
The duration time can be directly used as a translation input index after normalization treatment. If the index sensitivity needs to be improved, the system can also introduce a weighting mechanism to carry out weighting correction on the factors such as the state fluctuation amplitude, the fluctuation frequency or the smoothness of the cut-in path in different stages, so as to generate a more discernable input expression value.
In step S502, the system obtains a preset reference index, where the reference index is obtained based on a set of statistics of historical high quality translation behavior segments. The high-quality fragments are derived from behavior process data of other translators with high-score expression when translating tasks of the same type (namely, task types are consistent), and the translators generally have the characteristics of high response speed, stable behavior cut-in and good task completion evaluation, so that the high-quality fragments can be used as a reference template for evaluating the input state of the current translator. By counting the average time period from the task start to the concentration state in the group of high-quality fragments, the system obtains a preset reference index for comparison.
The second adjustment factor reflects the overall degree of deviation of the target translator between the current state and the efficient behavioral reference. If the value is positive, the translator usually takes more time to enter the translation state than the high-quality reference translator, the task response speed is slower, and if the value is negative, the translator can enter the concentration state more quickly, and the input efficiency is higher. By quantifying and averaging the deviation degrees, the general input performance of a target translator in different tasks recently can be effectively measured, objective basis is provided for the down-regulation or up-regulation of the scheduling priority, and the task allocation strategy is more robust and data driven.
Further, the data monitoring method based on image recognition further comprises the following steps:
Step S600, the original scheduling priority value is comprehensively corrected based on the first adjustment factor and the second adjustment factor.
Specifically, fig. 4 shows a flowchart for correcting the original scheduling priority value.
The method for comprehensively correcting the original scheduling priority value based on the first adjustment factor and the second adjustment factor specifically comprises the following steps:
Step S601, a preset priority value correction formula is called, and the original scheduling priority value is subjected to weighted correction by combining a first adjustment factor and a second adjustment factor to obtain a corrected scheduling priority value;
Step S602, the corrected scheduling priority value is applied to the task allocation flow of the target translator, so as to optimize the scheduling sequence or task matching decision of the target translator.
The priority value correction formula is as follows: wherein Refers to the revised scheduling priority value,Referring to the original scheduling priority value,Refers to the first adjustment factor, i.e., the average slope of the time series,Refers to the adjustment weight of the first adjustment factor,Refers to the total number of associated translated segments,Refers to the firstThe translation input index corresponding to each associated translation segment,Refers to the preset reference index of the index,Refers to the second adjustment factor, i.e. the average of all deviation magnitudes,Refers to the adjustment weight of the second adjustment factor, andAndAre all greater than 0.
In the embodiment of the invention, the first adjustment factor and the second adjustment factor are designed to participate in the correction of the scheduling priority value together, so that the method has obvious technical advantages and application value. The behavior dimensions measured by the two are not overlapped with each other, but have high correlation under the specific background of equipment switching, and the real adaptation capability of the translator can be reflected from the cooperation of the behavior trend and the instant state, so that a more complete and stable scheduling optimization mechanism with dynamic response capability is constructed.
First, the first adjustment factor reflects the adaptability variation trend of the translator in the switching scene of the plurality of historical equipment. The essence is that a time sequence is constructed through the equipment adaptation index, and the average slope of the time sequence is calculated, so that whether the behavior of a target translator is gradually stabilized, improved or degenerated in the similar equipment migration task is judged. Such historical sequence-based trend factors can reveal the impact of device switching on the long-term behavioral state of the translator. For example, for some translators, frequent switching from the mobile terminal to the desktop terminal may form a stable adaptation path, which may appear as a gradual increase in the adaptation index, while other translators may have a repetitive adjustment period after each switching, which may appear as a large fluctuation or even a downward slip in the adaptation index. The factor is introduced, so that the system is not only concerned with the current performance during scheduling, but also can recognize whether a translator has the capability of stabilizing the cross-equipment work in advance, thereby reducing task interruption or quality fluctuation caused by scheduling errors.
Second, the second adjustment factor is an assessment of the "cut-in speed" required by the translator in the current translation task from the start of the task to fully enter a steady translation state. The factor calculates the magnitude of investment deviation for the translator in different tasks by comparing with the average investment duration extracted from a set of high quality translation samples. The method does not depend on long-time historical trend, but focuses on the state response capability of the current task earlier stage, and is an important index for measuring the instant input efficiency of the translator. After the equipment is switched, the translator often needs a certain time to be re-adapted to technical environments such as a terminal interface, voice input, audible feedback and the like, so that the factor is very critical for judging whether the translator has completed state transition.
The two factors are selected as the basis of scheduling correction, and the device switching has the following two typical characteristics that the device switching has a certain degree of operation environment reconstruction and causes disturbance to the behaviors of a translator in a short period, and the disturbance can be reflected as local response delay in each task and can also accumulate to form habitual influence in long-term tasks. Thus, relying on only one view is not enough to fully reflect the translator's scheduling adaptation capability. If only the historical trend (the first adjustment factor) is relied on, the immediate risk of high-intensity tasks being just completed by the translator before a certain task and the state drop can be ignored, and if only the current input performance (the second adjustment factor) is relied on, the overall equipment adaptation stability of the translator can be ignored, and the frequent improper scheduling is caused. Therefore, the two factors are combined and used to form a double check logic of 'long-term behavior trend+current input efficiency', so that the trend prejudgment and the real-time state response balance can be realized in the dispatching process, and the reliability and the context adaptability of the dispatching behavior are effectively improved.
Furthermore, the double-factor mechanism is not only suitable for translator task scheduling scenes, but also has higher universality in other intelligent distribution systems with multi-terminal and multi-task switching, and can be widely applied to multiple fields of remote collaboration, virtual customer service, cross-equipment man-machine interaction and the like. Therefore, the dual-factor joint scheduling correction mechanism provided by the invention is especially suitable for the unique working background of equipment switching, and has obvious advantages in the aspects of depth of behavior understanding, system response efficiency, algorithm stability and the like.
The correction formula provided by the invention adopts a linear weighted exponential correction model, has the advantages of intuitiveness, adjustability, low calculation cost and the like, and is particularly suitable for embedded scheduling scenes and real-time system application. However, the present invention is not limited to the linear structure. According to the actual application requirements, the following alternative or expanded forms can be designed:
For example, based on the scale normalization process, the robustness to extreme values can be improved by adopting an exponential smoothing function or a piecewise nonlinear mapping mode, and in the situation of preference of trend driving or efficiency priority, a weight self-adaptive adjustment mechanism can be designed, namely AndThe method can dynamically update according to the historical performance, and can also introduce a threshold control mechanism to enable the correction behavior to be triggered only when the adjustment factor exceeds a set critical value, so as to avoid frequent adjustment in a fluctuation interval. In addition, if the system supports the embedding of the depth model, the priority correction function can be learned on the basis of a plurality of dimensional behavior data through a neural network or a regression model, so that more complex nonlinear dynamic allocation logic can be realized.
Further, fig. 5 shows an application architecture diagram of the system provided by the embodiment of the present invention.
In another preferred embodiment of the present invention, a data monitoring system based on image recognition includes:
The data obtaining module 100 is configured to obtain, when it is determined that the target translator has a specific device switching behavior, an original scheduling priority value, a history translation record, and a current image segment of the target translator, and determine a type of a current translation task.
Further, the data monitoring system based on image recognition further comprises:
the tag setting module 200 is configured to perform image recognition processing on the target translator based on the current image segment, and generate a current behavior tag of the target translator.
Further, the data monitoring system based on image recognition further comprises:
The data filtering module 300 is configured to filter a plurality of associated translation segments from the historical translation record according to the type of the current translation task, the current behavior tag, and the specific device switching behavior.
Further, the data monitoring system based on image recognition further comprises:
the first adjustment factor determining module 400 is configured to extract, based on the historical translation record, a device adaptation index of the target translator in each associated translation segment, and calculate and generate a first adjustment factor according to a time sequence variation trend of the device adaptation index.
Specifically, fig. 6 shows a block diagram of a first adjustment factor determining module 400 in the system according to an embodiment of the present invention.
In a preferred embodiment of the present invention, the first adjustment factor determining module 400 specifically includes:
An image segment extraction unit 401, configured to parse the history translation record, and extract an image behavior segment in a preset period after completing a specific device switching behavior in each associated translation segment by a target translator;
The device adaptation index generating unit 402 is configured to parse the image behavior segment based on an image recognition technology, extract behavior parameters of a face orientation, a lip movement rhythm, a gesture stability, and an eye movement track, calculate a convergence degree of each behavior parameter in a preset period, generate a corresponding behavior stability score, and weight and fuse each stability score to generate a device adaptation index;
the average slope calculating unit 403 is configured to construct a time sequence based on the device adaptation indexes corresponding to the plurality of associated translation segments, and take an average slope of the time sequence as a first adjustment factor.
Further, the data monitoring system based on image recognition further comprises:
The second adjustment factor determining module 500 is configured to intelligently evaluate each associated translation segment, generate a corresponding translation input index, obtain a preset reference index, and calculate to obtain a second adjustment factor based on a difference between each translation input index and the preset reference index.
Specifically, fig. 7 shows a block diagram of a second adjustment factor determining module 500 in the system according to an embodiment of the present invention.
In a preferred embodiment of the present invention, the second adjustment factor determining module 500 specifically includes:
the translation input index generating unit 501 is configured to parse each associated translation segment, monitor behavior of a target translator at a translation task initiation stage, identify a behavior change process that the target translator experiences from task initiation to entering a stable translation state, measure a duration time required for the target translator to reach a translation concentration state, and generate a corresponding translation input index based on the duration time;
A preset reference index obtaining unit 502, configured to obtain a preset reference index, where the preset reference index refers to an average input duration counted in a plurality of high-quality translation behavior segments;
the deviation amplitude calculating unit 503 is configured to calculate a deviation amplitude between each of the translation input indexes and a preset reference index, and take an average value of all the deviation amplitudes as a second adjustment factor.
Further, the data monitoring system based on image recognition further comprises:
The priority value correction module 600 is configured to comprehensively correct the original scheduling priority value based on the first adjustment factor and the second adjustment factor.
Specifically, fig. 8 shows a block diagram of a priority value correction module 600 in the system according to an embodiment of the present invention.
In a preferred embodiment of the present invention, the priority value correction module 600 specifically includes:
the priority value correction unit 601 is configured to invoke a preset priority value correction formula, and combine the first adjustment factor and the second adjustment factor to perform weighted correction on the original scheduling priority value, so as to obtain a corrected scheduling priority value;
the corrected priority value application unit 602 is configured to apply the corrected scheduling priority value to a task allocation flow of the target translator, and is configured to optimize a scheduling order or task matching decision of the target translator.
The priority value correction formula is as follows: wherein Refers to the revised scheduling priority value,Referring to the original scheduling priority value,Refers to the first adjustment factor, i.e., the average slope of the time series,Refers to the adjustment weight of the first adjustment factor,Refers to the total number of associated translated segments,Refers to the firstThe translation input index corresponding to each associated translation segment,Refers to the preset reference index of the index,Refers to the second adjustment factor, i.e. the average of all deviation magnitudes,Refers to the adjustment weight of the second adjustment factor, andAndAre all greater than 0.
It should be understood that, although the steps in the flowcharts of the embodiments of the present invention are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in various embodiments may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
Those skilled in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, where the program may be stored in a non-volatile computer readable storage medium, and where the program, when executed, may include processes in the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the invention and are described in detail herein without thereby limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (10)
1. A method for monitoring data based on image recognition, the method comprising:
when the fact that the target translator has specific equipment switching behaviors is determined, an original scheduling priority value, a historical translation record and a current image fragment of the target translator are obtained, and the type of a current translation task is determined;
based on the current image fragment, performing image recognition processing on the target translator to generate a current behavior label of the target translator;
According to the type of the current translation task, the current behavior label and the switching behavior of the specific equipment, a plurality of associated translation fragments are screened out from the historical translation record;
Based on the historical translation record, extracting the equipment adaptation index of the target translator in each associated translation segment, and calculating and generating a first adjustment factor according to the time sequence change trend of the equipment adaptation index;
performing intelligent evaluation on each associated translation segment, generating a corresponding translation input index, acquiring a preset reference index, and calculating to obtain a second adjustment factor based on the difference between each translation input index and the preset reference index;
And comprehensively correcting the original scheduling priority value based on the first adjustment factor and the second adjustment factor.
2. The method for monitoring data based on image recognition according to claim 1, wherein the step of extracting the device adaptation index of the target translator in each associated translation segment based on the history translation record, and generating the first adjustment factor according to the time-series trend calculation of the device adaptation index comprises:
analyzing the historical translation record, and extracting image behavior fragments of a target translator in a preset period after completing the switching behavior of specific equipment in each associated translation fragment;
Analyzing the image behavior segments based on an image recognition technology, extracting behavior parameters of face orientation, lip movement rhythm, gesture stability and eye movement track, calculating convergence degree of the behavior parameters in a preset period, generating corresponding behavior stability scores, and weighting and fusing the stability scores to generate equipment adaptation indexes;
and constructing a time sequence based on the device adaptation indexes corresponding to the associated translation fragments, and taking the average slope of the time sequence as a first adjustment factor.
3. The method for monitoring data based on image recognition according to claim 2, wherein the steps of intelligently evaluating each associated translation segment, generating a corresponding translation input index, obtaining a preset reference index, and calculating a second adjustment factor based on a difference between each translation input index and the preset reference index, comprise:
Analyzing each associated translation segment, performing behavior monitoring on a target translator at the translation task starting stage, identifying the behavior change process of the target translator from task starting to stable translation state entering, measuring the duration time required by the target translator to reach the translation concentration state, and generating a corresponding translation input index based on the duration time;
acquiring a preset reference index, wherein the preset reference index refers to average input duration counted in a plurality of high-quality translation behavior fragments;
and calculating the deviation amplitude value of each translation input index compared with the preset reference index, and taking the average value of all the deviation amplitude values as a second adjustment factor.
4. The image recognition-based data monitoring method of claim 3, wherein the step of comprehensively modifying the original scheduling priority value based on the first adjustment factor and the second adjustment factor comprises:
a preset priority value correction formula is called, and the original scheduling priority value is subjected to weighted correction by combining the first adjustment factor and the second adjustment factor to obtain a corrected scheduling priority value;
and applying the corrected scheduling priority value to a task allocation flow of the target translator, and optimizing the scheduling sequence or task matching decision of the target translator.
5. The image recognition-based data monitoring method of claim 4, wherein the priority value correction formula is: wherein Refers to the revised scheduling priority value,Referring to the original scheduling priority value,Refers to the first adjustment factor, i.e., the average slope of the time series,Refers to the adjustment weight of the first adjustment factor,Refers to the total number of associated translated segments,Refers to the firstThe translation input index corresponding to each associated translation segment,Refers to the preset reference index of the index,Refers to the second adjustment factor, i.e. the average of all deviation magnitudes,Refers to the adjustment weight of the second adjustment factor, andAndAre all greater than 0.
6. The data monitoring system based on image recognition is characterized by comprising a data acquisition module, a label setting module, a data screening module, a first adjustment factor determining module, a second adjustment factor determining module and a priority value correcting module, wherein:
The data acquisition module is used for acquiring an original scheduling priority value, a historical translation record and a current image fragment of the target translator when determining that the target translator has specific equipment switching behaviors, and determining the type of a current translation task;
The label setting module is used for carrying out image recognition processing on the target translator based on the current image fragment to generate a current behavior label of the target translator;
The data screening module is used for screening a plurality of associated translation fragments from the historical translation record according to the type of the current translation task, the current behavior label and the switching behavior of the specific equipment;
The first adjustment factor determining module is used for extracting the equipment adaptation index of the target translator in each associated translation segment based on the historical translation record, and calculating and generating a first adjustment factor according to the time sequence change trend of the equipment adaptation index;
The second adjustment factor determining module is used for intelligently evaluating each associated translation segment, generating a corresponding translation input index, acquiring a preset reference index, and calculating to obtain a second adjustment factor based on the difference between each translation input index and the preset reference index;
And the priority value correction module is used for comprehensively correcting the original scheduling priority value based on the first adjustment factor and the second adjustment factor.
7. The image recognition-based data monitoring system of claim 6, wherein the first adjustment factor determination module specifically comprises:
The image segment extraction unit is used for analyzing the history translation record and extracting image behavior segments in a preset period after the target translator completes the switching behavior of the specific equipment in each associated translation segment;
The device adaptation index generation unit is used for analyzing the image behavior segments based on the image recognition technology, extracting behavior parameters of face orientation, lip movement rhythm, gesture stability and eye movement track, calculating the convergence degree of each behavior parameter in a preset period, generating corresponding behavior stability scores, and weighting and fusing each stability score to generate a device adaptation index;
And the average slope calculation unit is used for constructing a time sequence based on the device adaptation indexes corresponding to the associated translation fragments, and taking the average slope of the time sequence as a first adjustment factor.
8. The image recognition-based data monitoring system of claim 7, wherein the second adjustment factor determination module specifically comprises:
The translation input index generation unit is used for analyzing each associated translation fragment, performing behavior monitoring on a target translator at the translation task starting stage, identifying the behavior change process of the target translator from task starting to stable translation state, measuring the duration time required by the target translator to reach the translation concentration state, and generating a corresponding translation input index based on the duration time;
the preset reference index acquisition unit is used for acquiring a preset reference index, wherein the preset reference index refers to average input time counted in a plurality of high-quality translation behavior fragments;
and the deviation amplitude calculation unit is used for calculating the deviation amplitude of each translation input index compared with the preset reference index, and taking the average value of all the deviation amplitudes as a second adjustment factor.
9. The image recognition-based data monitoring system of claim 8, wherein the priority value correction module specifically comprises:
the priority value correction unit is used for calling a preset priority value correction formula, and combining the first adjustment factor and the second adjustment factor to carry out weighted correction on the original scheduling priority value so as to obtain a corrected scheduling priority value;
And the corrected priority value application unit is used for applying the corrected dispatching priority value to the task allocation flow of the target translator and optimizing the dispatching sequence or task matching decision of the target translator.
10. The image recognition-based data monitoring system of claim 9, wherein the priority value correction formula is: wherein Refers to the revised scheduling priority value,Referring to the original scheduling priority value,Refers to the first adjustment factor, i.e., the average slope of the time series,Refers to the adjustment weight of the first adjustment factor,Refers to the total number of associated translated segments,Refers to the firstThe translation input index corresponding to each associated translation segment,Refers to the preset reference index of the index,Refers to the second adjustment factor, i.e. the average of all deviation magnitudes,Refers to the adjustment weight of the second adjustment factor, andAndAre all greater than 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510652444.1A CN120218561B (en) | 2025-05-21 | 2025-05-21 | Data monitoring method and system based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510652444.1A CN120218561B (en) | 2025-05-21 | 2025-05-21 | Data monitoring method and system based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120218561A true CN120218561A (en) | 2025-06-27 |
| CN120218561B CN120218561B (en) | 2025-09-19 |
Family
ID=96115628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510652444.1A Active CN120218561B (en) | 2025-05-21 | 2025-05-21 | Data monitoring method and system based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120218561B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2546064C1 (en) * | 2013-11-12 | 2015-04-10 | Общество с ограниченной ответственностью "Технологии управления переводом" | Distributed system and method of language translation |
| US20160162478A1 (en) * | 2014-11-25 | 2016-06-09 | Lionbridge Techologies, Inc. | Information technology platform for language translation and task management |
| CN112288211A (en) * | 2020-07-07 | 2021-01-29 | 南京易杰智信息科技有限公司 | Method, system and device for distributing translation tasks |
-
2025
- 2025-05-21 CN CN202510652444.1A patent/CN120218561B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2546064C1 (en) * | 2013-11-12 | 2015-04-10 | Общество с ограниченной ответственностью "Технологии управления переводом" | Distributed system and method of language translation |
| US20160162478A1 (en) * | 2014-11-25 | 2016-06-09 | Lionbridge Techologies, Inc. | Information technology platform for language translation and task management |
| CN112288211A (en) * | 2020-07-07 | 2021-01-29 | 南京易杰智信息科技有限公司 | Method, system and device for distributing translation tasks |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120218561B (en) | 2025-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI731297B (en) | Risk prediction method and apparatus, storage medium, and server | |
| CN119831100A (en) | Intelligent classroom teaching optimization method and system combining behavior recognition and Internet of things | |
| CN110309799B (en) | Camera-based speaking judgment method | |
| CN109472206A (en) | Methods of risk assessment, device, equipment and medium based on micro- expression | |
| CN109793526B (en) | Lie detection method, device, computer equipment and storage medium | |
| CN113571096B (en) | Speech emotion classification model training method and device, computer equipment and medium | |
| CN113435362A (en) | Abnormal behavior detection method and device, computer equipment and storage medium | |
| WO2008157811A1 (en) | Selective sampling of user state based on expected utility | |
| CN119884981A (en) | User preference analysis method based on memory-enhanced multi-mode large model | |
| CN120075203B (en) | Conference enhancement method, device, equipment and storage medium | |
| CN119676493B (en) | A method and system for user input information in an all-in-one smart television | |
| CN110930989A (en) | Speech intention recognition method and device, computer equipment and storage medium | |
| CN119919003A (en) | An expert review system and evaluation method for multi-dimensional intelligent evaluation | |
| CN117877445A (en) | Screen brightness adjusting method based on screen angle | |
| CN120218561B (en) | Data monitoring method and system based on image recognition | |
| CN114997373B (en) | Network depth dynamic adjustment method, device, computer equipment and storage medium | |
| CN120406746B (en) | An intelligent interactive control method and system for electronic photo frame based on eye tracking | |
| CN111950481A (en) | Facial expression quantification and recognition method in image and its device, equipment and medium | |
| CN119739280A (en) | Iris recognition personalized preference setting method and system in VR environment | |
| CN117896484A (en) | Target finding method, device, equipment and medium based on visual intercom system | |
| CN117151114A (en) | Intention recognition method and device | |
| CN121330748B (en) | Industrial and commercial management teaching model identification and feedback method and system based on machine vision | |
| CN121213086B (en) | A method, device and medium for intelligent customer service response in banks | |
| US20250378962A1 (en) | Information processing system and information processing method | |
| KR102927455B1 (en) | Waste Object Recognition System Using Hyperparameter Optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |