24results about How to "Address subjectivity" patented technology

The embodiment of the invention discloses a multi-mode-based after-examination illegal behavior judgment method under an open education system, and the method comprises the steps: obtaining a training data set related to after-examination behaviors, and carrying out the training of an after-examination illegal behavior judgment model through the training data set; predicting a new single image and / or an image-text pair by using the trained model to obtain a post-examination illegal behavior category; wherein in the training process, a single image sample is input into a Resnet network, and a first classification feature of the single image sample is obtained; inputting the image-text pair sample into the CLIP network to obtain a second classification feature of the image-text pair sample; fusing the first classification feature and the second classification feature, and inputting the fused feature into an MLP full connection layer to obtain a third classification feature; and enabling the third classification feature to pass through an activation function to obtain a post-examination illegal behavior category of the input data. According to the embodiment of the invention, the illegal behavior after examination is automatically identified through multi-modal fusion.

The invention discloses a traceability anti-counterfeiting management method for artistic paintings, and belongs to the technical field of traceability anti-counterfeiting, and the method specifically comprises the steps: endowing each artistic painting with a 12-bit unique serial number, and uploading the serial numbers to a cloud platform for management; the serial numbers are randomly divided into X groups of Y-bit numbers, and each group of numbers is mapped to gray values under different gray levels A, B and C according to own numerical values to be displayed; in the transportation process, positioning information of the transportation vehicle is collected in real time to generate a transportation track, and the transportation track is bound with the serial number and uploaded to the cloud platform; a purchaser can inquire a transportation track by inputting a serial number or upload a serial number photo to verify authenticity; the system judges whether the serial number display in the photo is consistent with the gray value display of the set value according to the division of the serial number and the gray mapping, so as to prompt a certified product or a counterfeited product; according to the invention, through a multi-verification mechanism, the anti-counterfeiting efficiency and accuracy of the artwork are effectively improved.

The present invention relates to a method for analyzing a job resource consumption pattern based on BWO and clustering algorithms, and belongs to the field of big data computing technology; the method comprises: (1) job operation and task scheduling; (2) monitoring of working node resource indicators; (3) data preprocessing and data set generation; (4) running a BWO / K-prototypes algorithm based on a White Whale optimization algorithm and a K-prototypes algorithm; (5) application of the optimal solution of the BWO / K-prototypes algorithm; (6) clustering result analysis; (7) acquisition and storage of the resource consumption pattern analysis results of a Flink job. The present invention can not only improve the execution efficiency of a Flink cluster and avoid exception throwing, but also reduce resource occupation and avoid unnecessary resource waste.

The application provides an animal immune evaluation method based on a Lutzomyia longipalpis infection model, and comprises the following steps: S1, experimental animal preparation and screening; S2, Lutzomyia longipalpis preparation and quality verification; S3, Lutzomyia longipalpis infection of experimental animals; S4, immune related index detection; S5, animal immune function evaluation; and S6, repeatability verification and data summarization. The application realizes systematic and objective evaluation of the effect of immune candidate drugs.

The application discloses a self-adaptive multi-modal data fusion bridge health monitoring method and system, generates a response sample database through establishment of a bridge nonlinear dynamics equation, pre-processes bridge sensor data and trains a deep learning model to extract damage features, and self-adaptively optimizes model parameters; based on model output, real-time and historical data are combined to perform bridge health evaluation, and a monitoring strategy is dynamically adjusted or early warning is triggered; an edge computing device is deployed to perform pre-processing and damage identification, and long-term trend analysis is realized in cooperation with a cloud platform; macro-structure damage and micro-crack data are fused through cross-scale feature extraction; an artificial intelligence algorithm is used to generate a bridge health evaluation result, a bridge maintenance and repair scheme and a decision-level early warning signal. The application combines nonlinear dynamics modeling, self-adaptive deep learning, edge cloud collaborative computing and cross-scale analysis, improves complex damage identification precision and monitoring intelligent level, and is suitable for large-scale traffic network bridge monitoring management.

The invention discloses a motor quality inspection method and computer equipment, and relates to the field of industrial automation and precision manufacturing, and the method comprises the steps: determining a plurality of part actual shapes and a plurality of surface actual coatings of a target motor according to image data; comparing the actual shapes of the plurality of parts with standard shapes of the parts corresponding to the plurality of target motor parts to obtain first quality inspection results of the actual shapes of the plurality of parts; component color features and component glossiness features are determined according to the multiple surface actual coatings; comparing the component color feature and the component glossiness feature with a component standard color and a component standard glossiness in the surface standard coating respectively to obtain a plurality of second quality inspection results of the surface actual coating; and when the first quality inspection result and the second quality inspection result meet the preset target motor quality standard, returning a quality inspection qualification result. By implementing the method, personalized quality inspection of different motor types is realized through image analysis, and the accuracy and reliability of quality inspection are improved.

The invention discloses a transportation equipment capability assessment method based on an improved neural network. The transport equipment capability assessment method based on the improved neural network comprises the following steps: firstly, constructing an index system influencing the transport capability, and determining an index importance degree by using an improved neural network algorithm; secondly, determining a membership function and an evaluation result based on a fuzzy theory; and finally, verifying the effectiveness of the evaluation method through experimental simulation and comparative analysis. The method provided by the invention plays an important role in improving the capacity of the transportation equipment and guiding the construction and development of the transportation equipment.

The application discloses a coating defect diagnosis method and device based on physical mechanism constraint, and relates to the technical field of coating detection. The method comprises the following steps: synchronously acquiring multi-modal detection data in the same detection area of a to-be-detected coating, wherein the multi-modal detection data at least comprises hyperspectral data, Fourier transform infrared spectrum data and electrochemical impedance spectrum data; extracting optical characteristics representing optical response of the coating, chemical characteristics representing chemical composition of the coating and electrochemical characteristics representing electrochemical behavior of the coating interface from the multi-modal detection data; inputting the optical characteristics, the chemical characteristics and the electrochemical characteristics into a pre-trained feature coupling model based on physical mechanism constraint; wherein the coating degradation physical mechanism is embedded into the training process of the feature coupling model as a constraint condition, and the multi-modal features are coupled and modeled; and outputting quantitative diagnosis parameters of the coating defect through the feature coupling model. The application realizes accurate and comprehensive coating diagnosis based on the coupling model.

The invention discloses an artificial intelligence medical image quality control method applied to clinical images, and relates to the technical field of medical image quality control. The artificial intelligence medical image quality control method applied to the clinical image comprises the following five steps: improving on the basis of a conventional word sequence training model by using a BiLSTM (Bidirectional Circulating Long Short Term Memory) neural network sequence prediction model, and improving input data of sequence prediction from a mode of predicting a next word by word input into a mode of predicting a next word by word input. According to the method, a sentence is input according to sentences, learning, recognition and prediction of a combination sequence between characters are carried out in a mode of predicting a word sequence, a model is trained in a mode of simulating word sequence prediction by taking the sentences as a whole, and the problem that difference between different prediction report groups cannot be unifiedly applicable to error screening index thresholds in practical application is solved; the occurrence of misdiagnosis is reduced, and the efficiency and accuracy of quality control marking are improved.

The invention discloses a photovoltaic sunflower based on a photoresistor, and relates to the technical field of photovoltaic building integration, the photovoltaic sunflower comprises a photovoltaic panel, a cleaning robot and an automatic dust removal module, the top of the outer wall of the photovoltaic panel is movably provided with the cleaning robot, and the cleaning robot is provided with the automatic dust removal module; the automatic dust removal module is used for cleaning dust on the surface of the photovoltaic panel; and the automatic dust removal module comprises a dust removal brush, a first rotating shaft, a first motor, a first rotating shaft and a first roller. By installing the automatic dust removal module, the automatic dust removal function of the photovoltaic panel is achieved, the problem that dust affects the photoelectric conversion efficiency of the photovoltaic panel is solved, the power generation efficiency of the photovoltaic sunflower is improved, and dependence on manual labor force is reduced.

This invention discloses a big data analysis-based method and system for art teaching management, relating to the field of educational information technology. The method includes: acquiring multi-source teaching big data for a target art teaching unit; constructing a pattern graph extraction unit and a teaching pattern graph based on the multi-source teaching big data; collecting stage-specific student assignment data and stage-specific student process data for the target art teaching unit according to preset stage window constraints; extracting pattern graphs from the stage-specific student assignment data through the pattern graph extraction unit to obtain a student pattern graph set; comparing the student pattern graph set with the teaching pattern graph; making evaluation decisions based on the comparison results; and implementing stage-specific management of the target art teaching unit based on the evaluation decision report. This invention solves the problems in existing teaching management where teacher evaluation is subjective and difficult to quantify, teaching evaluation lacks process analysis, and lacks the ability to deeply analyze the content of assignments themselves.

The invention specifically relates to a metal oxide arrester leakage current on-line monitoring system and method, and relates to the technical field of power equipment monitoring, and the system comprises a multi-physical field sensing array module; a synchronization and data aggregation gateway module; the central analysis server is connected with the fusion module; and a multi-dimensional state evaluation and decision module. According to the invention, through the collaborative design of the electrical, acoustic and thermal three-dimensional sensing units, the problems of early defect misjudgment and interference signal misjudgment caused by the fact that traditional MOA monitoring only depends on power frequency parameters are solved; nanosecond-level partial discharge pulses are captured in the electrical dimension; the acoustic dimension realizes three-dimensional positioning of defects; the thermal dimension is combined with an environment compensation algorithm, so that tiny temperature rise can be identified; and in cooperation with triple cross validation of a space-time correlation algorithm, internal defects and external interference are effectively distinguished.

This invention discloses a method and system for detecting superficial fungi using microscopic images based on the YOLO model. The method involves collecting and annotating microscopic images of clinical superficial fungal disease samples. The annotated images are then classified and input into deep learning-based target detection models (including SSD-VGG16, Faster R-CNN, YOLOv5, YOLOv8, and YOLOv11 models) for training and testing. The system adjusts the Intersection over Union (IOU) threshold and evaluates the effectiveness of the confidence aggregation strategy. After preliminary performance evaluation, the best-performing model, YOLOv11, is optimized. The IOU threshold is optimized, and the system evaluates thresholds within the range of 0.25-0.75. Image-level confidence analysis is performed to evaluate the aggregation strategy. The YOLOv11 model of this invention performs optimally at 400× magnification, with an F1 score of 86.60% for hyphae detection and 95.90% for spore detection.

The invention provides a vehicle unmanned driving system decision-making situation assessment method, device and equipment and a medium, and relates to the field of intelligent driving, and the method comprises the steps: obtaining an environment element set and a decision-making target set facing a vehicle unmanned driving system; determining the influence of the environment element set on different sub-levels formed by the decision target set to obtain a first determination degree, a second determination degree and a third determination degree; carrying out weight distribution on the first determination degree, the second determination degree and the third determination degree to obtain a target determination degree so as to balance subjective and objective differences; drawing a comprehensive situation map by taking time as a horizontal coordinate based on the comprehensive situation represented by the target certainty degree; and decision planning is carried out. According to the invention, based on multi-target value consistency measurement, differential cognition and fuzziness among different targets can be eliminated on a cognition level, and the problem that subjective preference and cognition of the vehicle unmanned driving system in a complex decision-making environment are inconsistent is effectively solved.

The invention discloses a sinter drum strength prediction model based on an improved Transform network and a system of the sinter drum strength prediction model. The model firstly obtains continuous state data of a sintering process, constructs a sintering process variable matrix M1 and a variable variation trend matrix M2 after preprocessing, constructs an improved Transform network model composed of a variable sequence feature extraction branch-regression block and a variable sequence variation trend feature extraction branch-regression block by adopting an Encoder structure of Transform as a basic framework, and finally constructs a variable sequence variation trend feature extraction branch-regression block. And inputting the matrix M1 and the matrix M2 into an improved Transform network model, carrying out weighted fusion on feature variables extracted from the two branches, inputting a fused variable Z as a regression block, and outputting a predicted value. According to the model, a deep learning network is adopted, and data features are extracted through operations such as a self-attention mechanism and residual connection, so that process data change trend information loss is avoided, the prediction result accuracy and the prediction speed are greatly improved, and the technical problems of time lag, manual subjectivity and the like in an existing detection method are effectively solved.

The invention belongs to the technical field of intelligent analysis and early warning, and discloses a traffic management public opinion intelligent analysis and early warning system based on VIT multi-mode, which comprises a platform adaptive data acquisition unit, a VIT cross-mode public opinion situation awareness module, a dynamic weight public opinion risk assessment module, a knowledge mapping public opinion guide aid decision module and a grading early warning push unit. According to the platform adaptive data acquisition unit, adaptive strategies are designed for various short video platforms, timed incremental pulling and high-heat real-time triggering acquisition are combined, videos, metadata and text information are covered at the same time, and the problems that existing public opinion monitoring is limited in coverage range and low in efficiency are solved; multi-modal analysis and association fusion are achieved through the VIT cross-modal public opinion situation awareness module, a three-level index system and real-time weight adjustment of the dynamic weight public opinion risk assessment module are matched, artificial experience judgment is replaced with quantitative calculation, and the problems that public opinion risk assessment is high in subjectivity and non-uniform in standard are solved.

The invention belongs to but is not limited to the technical field of epileptic focus positioning, and discloses an epilepsy lesion quantitative feature selection method and system based on 18F-FDG PET imaging, and the method comprises the steps: collecting 18F-FDG PET imaging data of a patient through an image collection module; the image preprocessing module is used for preprocessing an original PET image; the feature extraction module is used for extracting potential features of epilepsy lesions by using multiple algorithms; the feature selection module performs feature selection by using a machine learning algorithm; the epileptic focus positioning and analysis module analyzes the PET data of the patient through a classification model in combination with the selected features, generates a probability heat map of the epileptic focus and displays the position of the epileptic focus; and the result display and report module visualizes the analysis result and generates a detailed analysis report. According to the method, the limitation of subjectivity and single-dimensional analysis in the prior art is effectively solved, and the positioning precision and clinical application value of the epilepsy lesion are remarkably improved.

The invention discloses an energy consumption dynamic allocation method based on hospital department use time and regional area, and relates to the technical field of energy consumption dynamic allocation, and the method comprises the following working steps: S1, energy consumption department allocation and disassembly: determining the relationship between energy consumption and each department, the energy consumption can be generated by one or more departments, and the energy consumption can be generated by one or more departments; the apportionment mode is determined according to the number of participating departments; and S2, department time configuration: setting work time of the department, including daily work and rest, overtime and rest time, so as to judge whether the department participates in generation and distribution of energy consumption in a specific time period. The invention provides an objective, complete, reasonable and scientific energy consumption dynamic allocation mechanism, the accurate and scientific allocation of energy consumption is realized, and the subjectivity and one-sidedness of a traditional method are solved; department energy consumption composition analysis is generated, optimization of department energy consumption management is facilitated, and unreasonable energy consumption is reduced; and through association of the energy consumption and the performance indicators, resource investment is optimized, and the energy consumption use efficiency is improved.

The invention provides an automatic lifting and rocking method and system for converter tapping based on machine vision and a medium, and the method comprises the following steps: step 1, when judging that a slag blocking condition is met, triggering an infrared industrial camera to start to continuously collect an image sequence in a steelmaking process in real time; 2, an image preprocessing unit carries out filtering and noise reduction preprocessing on the obtained image sequence, the steel flow and the background are separated after preprocessing, and interference of a background area is eliminated; 3, an image feature extraction unit extracts feature parameters of the steel flow and calculates the proportion of the steel slag in the steel flow; whether slag stopping succeeds or not is judged through feature fusion of the slag stopping judgment unit; and 4, the prediction model in the online prediction unit receives the tapping parameters to predict the furnace lifting opportunity, a prediction result is transmitted to the converter control unit, and furnace shaking in advance when slag stopping fails or furnace lifting operation when slag stopping succeeds is automatically triggered. Real-time monitoring and automatic judgment in the slag stopping process are achieved, and the accuracy and reliability of judgment are improved. Meanwhile, the furnace lifting opportunity is dynamically optimized according to the slag stopping result, the steel making efficiency is improved while the molten steel quality is guaranteed, and unnecessary energy consumption and material waste are reduced.

The application discloses a kind of mineral resources reserves estimation method using fractal algorithm, step (S1): resource distribution map generation based on machine learning;Step (S2): feature matching mine site identification based on fractal geometry;Step (S3): reserve estimation based on regression analysis.The application is a new mineral resources reserves estimation calculation scheme used in the field of intelligent technology of mineral exploration.The scheme uses the "fractal" method in mathematics, combined with AI technology, to provide a complete solution for resource reserves calculation of target mine, especially weathered bauxite, weathered laterite nickel ore and other types of near-surface mine, to achieve the task goal of calculating the mineral resources reserves of target mining area without field exploration data or only using a small amount of artificial exploration data.

This invention relates to the field of environmental protection and resource recycling technology, and discloses a method for reducing the volume of oily emulsified wastewater. The method includes: real-time acquisition of multimodal sensor data such as turbidity, conductivity, oil-water interface height, pH, and temperature; and fusing and calculating a dynamic characteristic state vector composed of coalescence efficiency, phase separation rate, and interface stability. This invention, through the fusion of multimodal sensor data and deep feature extraction, transforms the monitoring of the demulsification process from macroscopic, qualitative subjective observation to microscopic, quantitative dynamic state characterization, achieving precise, objective, and real-time understanding of the process state. This fundamentally solves the problems of subjectivity and lag in manual judgment; it can predict the evolution trend of the system state over a future period, thereby transforming the control strategy from passive responsive regulation to proactive predictive intervention.

The application provides a fabric pilling automatic grading method and device, and relates to the technical field of textile detection. The method comprises the following steps: collecting a thermal infrared polarization image sequence of a fabric sample by using a thermal infrared polarization camera; calculating a polarization degree and a thermal decay time constant based on the sequence data, and extracting a micro-thermal radiation feature; simultaneously collecting a visible light image by using a visible light camera, and extracting an image feature; fusing the micro-thermal radiation feature and the image feature to generate multi-modal feature data; and finally identifying and determining a pilling grade by using a neural network model. By combining thermal physical characteristics and visual features, the problems of strong subjectivity of artificial grading and large limitation of single optical detection are overcome, and objective, accurate and efficient automatic grading is realized.

The application discloses a tower crane bolt axial force intelligent ultrasonic detection method and system based on multi-modal data fusion. First, a multi-level influence factor system is constructed based on fuzzy hierarchy comprehensive evaluation theory, and a high-risk area identification model is established. Based on the model, the regional risk value is calculated, and the monitoring distribution scheme is determined. In the selected area, a heterogeneous network composed of fiber Bragg grating strain sensors and ultrasonic sensors is arranged, and strain and ultrasonic data are synchronously collected. The multi-modal data are input into a deep belief network for feature fusion, and dynamic evaluation of the health state is realized through a fuzzy recognition model. Based on the dynamic grading evaluation result, an improved rain flow counting method and cumulative damage theory are used for fatigue life prediction. Finally, data fusion display and intelligent early warning are realized through a remote monitoring platform. The application realizes intelligent health management of the whole chain from risk identification, real-time monitoring to life prediction.

The embodiment of the present invention discloses a method for determining post-exam violation behavior based on multimodality under an open education system, including: obtaining a training data set about post-exam behavior, and using the training data set to train a post-exam violation behavior determination model; using the trained model to predict new single images and / or image-text pairs to obtain post-exam violation behavior categories; wherein, during the training process, a single image sample is input into a Resnet network to obtain a first classification feature of the single image sample; an image-text pair sample is input into a CLIP network to obtain a second classification feature of the image-text pair sample; the first classification feature is fused with the second classification feature, and the fused feature is input into an MLP fully connected layer to obtain a third classification feature; the third classification feature is passed through an activation function to obtain the post-exam violation behavior category of the input data. This embodiment automatically identifies post-exam violation behavior through multimodal fusion.