30results about How to "Reduce feature redundancy" patented technology

The invention relates to the technical field of image recognition, and discloses an automatic image recognition garbage classification whole process implementation method based on AI calculation. The method comprises the following steps: carrying out multi-modal data acquisition on to-be-identified garbage to obtain an image data matrix, a weight data vector and a spectral data matrix; performing semantic feature and detail feature extraction on the image data matrix to obtain a fusion feature map; performing channel attention and space attention mechanism enhancement to obtain an enhanced feature map; overlapping degree calculation and region merging are carried out to obtain garbage positioning data; calculating probability density distribution data of garbage category nodes based on the garbage positioning data, the weight data vector and the spectral data matrix; according to the method, the problems that the target size is small and the features are not obvious are solved, and the reliability of the classification result is ensured.

The invention relates to the technical field of data feature mining, in particular to a remote nursing management method based on data feature mining. The method comprises the following steps: acquiring specific conditions and demand data of a patient; performing data source selection and configuration according to the specific condition and demand data of the patient to obtain a data source configuration table; performing multi-dimensional data acquisition according to the data source configuration table, and performing multi-dimensional data preprocessing to obtain patient comprehensive data; carrying out individualized baseline modeling on the patient comprehensive data to obtain an individual baseline model; performing real-time data standardization on the patient comprehensive data to obtain standardized real-time data; performing deviation calculation according to the standardized real-time data and the individual baseline model to obtain a deviation value sequence; and performing index correlation analysis on the deviation value sequence to obtain an index correlation matrix. According to the invention, the utilization of data feature mining is improved, so that remote nursing management is more accurate and personalized, and finally the nursing quality and efficiency are improved.

The invention discloses a rail transit key equipment fault prediction method and system, and the method comprises the steps: carrying out the data collection and marking, and preliminarily calibrating an initial data state credibility coefficient; initializing and defining a feature set representing the health state of the equipment, associating abnormal item points with features, preliminarily calibrating correlation coefficients of the abnormal item points and the features, and performing initial feature extraction; establishing a fault prediction model by adopting a machine learning algorithm; using the established fault prediction model to diagnose main line operation data, outputting an online prediction result, performing data mark correction based on the result, and updating an abnormal item point and feature correlation coefficient and a corresponding initial data state credibility coefficient; and feature updating is carried out based on the corrected mark data, so that the fault prediction model is further updated and trained. According to the method, automatic data marking and data marking updating can be realized, online automatic model updating is realized in combination with a prediction result and actual maintenance feedback, and the model performance is further improved.

The invention discloses a physiological signal processing method and device and a medium, and is applied to the technical field of physiological signal processing. Processing the surface length sequence video data to obtain spatial-temporal characteristics of the surface length sequence video data; according to the spatial-temporal features, initial fast features and initial slow features are obtained, and the initial fast features and the initial slow features are features representing different time dimensions; according to the initial fast features, a time sequence difference module is combined to obtain deep fast features, and the time sequence difference module is used for carrying out feature extraction based on a preset state space model; according to the initial fast feature and the initial slow feature, combining a time sequence difference module to obtain a deep slow feature; and obtaining a physiological signal according to the deep-layer fast feature and the deep-layer slow feature. According to the application, the accuracy and reliability of the physiological signal can be effectively improved.

The invention discloses a PCB surface defect detection method based on improved YOLOv8, and belongs to the technical field of deep learning and quality detection, and the method comprises the following steps: constructing an improved YOLOv8 model based on an RFCBAMCConv module, an ScConv module and a VoVGSCSPC module; and inputting the PCB into the improved YOLOv8 model to obtain a PCB surface defect detection result.

The application discloses a cross-platform operation and maintenance replay scene intelligent identification method and a cross-platform operation and maintenance replay method. The identification method comprises the following steps: collecting operation and maintenance associated data from different platforms in real time, screening target associated feature data from the operation and maintenance associated data after preprocessing, determining time sequence features of the target associated feature data, and creating new features representing potential structures of the data; extracting key information from all the target associated feature data and the new features; extracting multi-scale features from the key information, performing feature fusion on the multi-scale features, and obtaining a plurality of composite features; screening a composite feature with the most independent features related to an identification target operation and maintenance scene from the plurality of composite features, and marking the composite feature as a target feature; and inputting the target feature into a target scene intelligent identification model to obtain a replay scene identification result. The method can automatically, efficiently and accurately identify an operation and maintenance operation scene that can be repeatedly executed between different operating systems and platforms.

The invention discloses an energy storage power station fault classification method based on local and global features of a relaxation stage. The method comprises the steps of 1, collecting terminal voltage, current and temperature data of the relaxation stage after charging and discharging in an actual working condition; 2, extracting local features and global features for fault detection from the collected terminal voltage, current and temperature features; 3, selecting a feature subset finally used for fault detection by using a filtering method and a packaging method in feature selection; and 4, battery cluster fault discrimination is carried out based on a classification algorithm random forest, KNN, logistic regression and an integrated voting strategy.

The application discloses a remote sensing change detection method based on a frequency domain representation decoupling network and belongs to the technical field of remote sensing image detection. The method comprises the following steps: acquiring double-time-phase remote sensing images, and pre-processing the double-time-phase remote sensing images to obtain double-time-phase pre-processed remote sensing images; constructing a remote sensing image change detection network based on frequency perception and invariant representation decoupling, and inputting the double-time-phase pre-processed remote sensing images into the remote sensing image change detection network to obtain a change detection result; the remote sensing image change detection network comprises a twin encoder, a frequency perception decoupling module and an invariant representation decoupling module; the change detection result is subjected to network parameter optimization using a loss function to obtain an optimal remote sensing image change detection network; and the optimal change detection result is acquired according to the optimal remote sensing image change detection network. The application further discloses an electronic device and a storage medium for implementing the above method.

The invention discloses a method and a system for detecting rice bacterial leaf blight based on unmanned aerial vehicle spatial spectrum fusion. The method comprises the following steps: firstly, constructing a rice bacterial leaf blight detection data set; a rice bacterial leaf blight severity grading detection double-branch model is constructed, and the model adopts a spectrum-space double-branch parallel coding and fusion framework and comprises a spectrum branch module, a space branch module and a cross attention fusion mechanism module. And inputting the complete image of the rice field into the trained rice bacterial leaf blight severity grading detection double-branch model, and obtaining severity grading detection results of the bacterial leaf blight in different rice growth periods of the target field. According to the invention, deep complementation and dynamic information balance of spectrum and spatial features of the multispectral unmanned aerial vehicle are realized, so that the accuracy of disease detection and the generalization ability of the model are effectively improved.

The invention discloses a real-time semantic segmentation method for attention-guided multi-scale information fusion, and the method comprises the following steps: 1, constructing a multi-scale feature aggregation module, and aggregating multi-scale detail information; 2, constructing an attention guiding feature correction module, and correcting information extracted by different branches to promote fusion of multi-scale information and local information; and 3, constructing an attention guiding feature reduction module, fusing feature maps extracted at different depths in the network, and making up image information lost in a feature transmission process. And 4, building an image segmentation network based on the three modules, efficiently extracting important features in the image, and accurately segmenting the complex scene image through a deep learning mode. A novel real-time scene semantic segmentation model is designed on the basis of an encoder-decoder architecture, the model precision and the model speed are better balanced, and under the conditions of low parameter and low calculation cost, accurate segmentation of image objects can be achieved, and meanwhile real-time performance is achieved.

The invention relates to the technical field of agricultural remote sensing monitoring, in particular to a method and system for measuring the number (area) of plants in a tobacco field based on visible light images of an unmanned aerial vehicle, and is suitable for automatic and high-precision growth monitoring and resource accounting of a large-scale tobacco field. Comprising six steps of tobacco field visible light image acquisition, image preprocessing, tobacco field region segmentation, tobacco field plant target detection and plant number statistics, tobacco field area calculation, and result post-processing and output, and through combination of an improved target detection algorithm and an image segmentation technology, high precision of tobacco field plant number statistics and high accuracy of area determination are realized. Meanwhile, the detection efficiency is guaranteed, the actual requirement of large-scale tobacco field monitoring is met, data acquisition, processing, analysis and output can be automatically completed in the whole process, manual intervention is not needed, the monitoring efficiency is greatly improved, and the labor cost is reduced.

The present invention relates to the field of intelligent transportation, and provides a method and device for processing geographic information data, wherein the method includes: obtaining geographic information data of a target area within a predetermined time range; calculating multiple data sets of a traffic feature set and target feature values ​​of each data set based on the geographic information data; using a predetermined physical knowledge base, once screening the features in the traffic feature set to obtain a spare traffic feature set, the physical knowledge base includes multiple feature sets with physical association relationships; based on the strong correlation between the features in the spare traffic feature set and the strong correlation between the features in the spare traffic feature set and the target features, performing a secondary screening on the features in the spare traffic feature set to obtain an optimal traffic feature set. The present invention can comprehensively and quickly determine the optimal traffic feature set that affects the target features, and is suitable for scenarios where computing resources on a satellite are limited.

The invention discloses a compound physical property prediction method, system, equipment and medium, and relates to the technical field of data driving, and the method comprises the steps: inputting the molecular characteristics of a compound into a physical property prediction model, and obtaining the correlation distance between the molecular characteristics based on a correlation distance clustering module, grouping the features and reducing linear redundant variables for linear fitting; the feature subspace combination module optimized based on the genetic algorithm performs feature combination global search and directional optimization by using the genetic algorithm so as to discover and evaluate the optimal combination among the features; extracting original features and physicochemical descriptors from the molecular structure by using a module suitable for linear and nonlinear coding, and searching features suitable for nonlinear fitting and linear fitting; according to the method, the correlation between a prediction result and a target variable is remarkably improved, so that the method shows superiority in the aspects of small sample physical property prediction and microscopic and macroscopic attributes of a multi-component complex system.

This invention discloses a method and device for diagnosing the health status of flow batteries based on particle swarm optimization (PSO), aiming to solve the problems of difficult modeling and poor real-time performance of traditional physical model-driven methods, and the unscientific feature selection and insufficient prediction stability of existing data-driven methods. The system is data-driven at its core, collecting multi-dimensional data such as voltage, current, and capacity of the flow battery during operation through sensors. After preprocessing, LASSO and grey relational analysis are used to jointly screen the optimal subset of health features. An extreme learning machine (ELM) is built as the main model for SOH prediction, using the ReLU activation function to improve nonlinear expression capabilities, and introducing a particle swarm optimization algorithm to optimize the weights and biases of the ELM, improving the instability caused by random initialization.

The invention relates to the technical field of oil well fracturing prediction, and solves the technical problems in the prior art that distribution of important features in fracturing data cannot be fully reserved generally through simple data segmentation, and low result accuracy in fracturing state prediction is easily caused. In particular to a fracturing state prediction system and method applying a machine learning intelligent decision-making system, and the method comprises the following steps: S1, obtaining original data of a gulongshale oil fracturing detection system, preprocessing the original data, and obtaining a unified table through an SQLJOIN method; the statistical feature extraction algorithm and the hierarchical clustering algorithm are used to construct a more excellent algorithm model, the accuracy and robustness of feature expression are improved by realizing feature extraction and clustering, the feature redundancy is reduced by combining statistical features and hierarchical clustering, the interpretability and prediction precision of a subsequent model are improved, and the prediction efficiency is improved. And meanwhile, the nonlinear relationship in the gulonium shale oil data is optimized, so that the subsequent steps are more accurate and smoother.

The invention discloses an obstetrical anti-phospholipid syndrome risk prediction method and system, and relates to the technical field of medical information processing, and the method comprises the steps: obtaining clinical sign data, laboratory specific antibody index data, past medical history data and diagnosis results of pregnant and lying-in women, and carrying out the preprocessing to obtain a standardized data set; extracting time sequence combination features and interaction features to form a fusion feature set; calculating a feature correlation intensity value and screening a core feature subset; extracting training features by using the core feature subset, and training a risk prediction model; extracting prediction features of a pregnant and lying-in woman to be predicted, and performing weight adjustment according to the pregnancy stage; and inputting the weighted features into a model to generate a risk score, determining a risk level and outputting early warning information. According to the invention, through multi-dimensional data analysis and dynamic weight adjustment, accurate prediction and timely early warning of the obstetrical anti-phospholipid syndrome risk are realized.

The present invention relates to a lightweight encrypted traffic analysis method and system based on a distillation representation model, comprising: obtaining an original data set of encrypted network traffic, performing unified processing on the data in the original data set and then dividing it according to a set ratio to generate various sample data sets; inputting pre-trained samples into a BERT model based on representation learning for pre-training to obtain a pre-trained model, inputting fine-tuning samples into the pre-training model for detailed task training to obtain a traffic classification teacher model; using a knowledge distillation method of multi-layer jump fusion learning to extract teacher information of the traffic classification teacher model, and inputting fine-tuning samples into a lightweight traffic analysis model for training; inputting test samples into the lightweight traffic analysis model to realize end-to-end target network encrypted traffic classification tasks; while ensuring classification accuracy, reducing analysis resource requirements, and facilitating the realization of online low-consumption encrypted traffic classification functions in delay-sensitive resource-constrained network scenario nodes.

The present application belongs to the field of human-computer interaction, and relates to a human-computer trust real-time measurement method and system based on heart rate variability. The present application performs principal component analysis on multiple time domain, frequency domain and nonlinear heart rate variability characteristics, and constructs a human-computer trust real-time measurement model on this basis; in application, user electrocardio signals are collected, heart rate variability characteristics are extracted from the user electrocardio signals; the heart rate variability characteristics are input into a load matrix, and a principal component matrix is calculated according to the load matrix; the principal component matrix is input into the human-computer trust real-time measurement model, and human-computer trust real-time measurement results are obtained. The present application is based on user heart rate variability characteristics, can realize real-time measurement of human-computer trust, and overcomes defects in measurement cost, accuracy, stability and anti-interference.

The invention belongs to the technical field of disease diagnosis based on gait analysis, and discloses a gait analysis disease diagnosis method and system based on feature interaction rebalance, and the method comprises the steps: carrying out the feature extraction of gait data, obtaining a local feature and a global feature, carrying out the cross enhancement of the local feature and the global feature, and carrying out the splicing operation of a result. Performing classification prediction by using a splicing result to obtain a classification result, calculating cross entropy loss between the classification result and a real label, calculating prototype cross entropy loss between the weighted local feature and a prototype vector as well as between the weighted global feature and the prototype vector, and calculating similarity loss between a similarity matrix and a transpose matrix thereof and the real label; and performing weighted combination on the loss to obtain a total loss function of the model, and training the model. According to the method, the intra-joint coupling relation and the inter-joint coupling relation are learned at the same time, cross-domain feature distribution differences are reduced by aligning cross-domain features, meanwhile, the fitting speed of a cross-domain extractor is balanced, and effective and objective gait-based reference is provided for disease diagnosis.

The application discloses a collaborative filtering recommendation method based on hypergraph generative multi-granularity contrast learning. The method comprises the following steps: obtaining historical interaction data of users and items, and constructing initial node embedding; adaptively constructing a hypergraph structure through a dynamic hypergraph constructor, and aggregating high-order collaborative signals by using a hypergraph convolution propagation mechanism; based on the enhanced node representation, a graph variational autoencoder is used to generate node-specific contrast views for each node; multi-granularity contrast learning is performed, including node-level contrast learning for maximizing the consistency of the same node under different views, and feature-level contrast learning for reducing feature redundancy and improving diversity; the recommendation task and the contrast learning loss are jointly optimized, and the model is trained to generate a recommendation list. Through dynamic hypergraph modeling of high-order correlation, generative contrast views and multi-granularity learning, the application effectively improves the accuracy, robustness and generalization ability of the recommendation, and can be widely applied to personalized scenarios such as e-commerce and content recommendation.

This application discloses a sound source localization model training method, a sound source localization method, and an apparatus, belonging to the field of sound signal processing technology. This application is applied to sound source localization scenarios, such as predicting the spatial position of a loosely connected component, wherein the component is mounted on a mechanical connection structure. This application uses sound signals as the data processing object, trains a sound source localization model based on pre-built training samples, and achieves automatic sound source localization based on the trained model. Since this method requires no manual intervention, compared to manual inspection, it not only saves labor costs and avoids the risks of working at heights, but is also more efficient, enabling real-time monitoring of the fastening status of the connected component.

The invention discloses a certificate type identification method and device, equipment and a medium, and relates to the field of image identification, and the key points of the technical scheme are as follows: obtaining a non-certificate image data set and a certificate image data set with a plurality of certificate types; a non-certificate image of one batch size of a non-certificate image data set and a certificate image of two batch sizes of a certificate image data set are taken as training data each time, and certificate categories to which images of each batch of certificate image data belong are different; the categories of the certificate images at the corresponding positions of the first batch and the second batch are the same, and the certificate images at the corresponding positions of the first batch and the second batch are different; training a pre-constructed certificate type identification model through the training data until the loss function converges, and obtaining a trained certificate type identification model; and identifying the to-be-identified certificate image based on the trained certificate type identification model to obtain an identification result of the certificate type.

The invention relates to the technical field of power operation and maintenance, and particularly discloses an insulator self-explosion defect detection method based on a lightweight adaptive YOLO model, and the method comprises the steps: designing a lightweight network architecture, achieving feature fusion, optimizing gradient shunting, constructing a decoupling head network, training a model, carrying out experimental verification and optimization, and applying the model. According to the insulator self-explosion defect detection method based on the lightweight adaptive YOLO model provided by the invention, the detection precision of small-target insulator defects can be improved, the calculation complexity of the model can be reduced, and the method adapts to the requirements of real-time performance and resource limitation in practical application.

1. Invention name: Time series prediction feature selection method and system based on partial autocorrelation function (PACF) and mutual information (MI) dynamic cooperation 2. Technical field: The present invention belongs to the technical field of time series prediction, and relates to a method and system for dynamically screening linear and nonlinear features, which are suitable for industrial equipment predictive maintenance, energy scheduling, financial risk control and other scenes. 3. Technical scheme: By detecting the non-stationarity (ADF test) and memory (Hurst index) of time series data, the weight parameter α of PACF and MI is dynamically adjusted (non-stationary data α∈[0.2, 0.4], strong memory data α∈[0.6, 0.8]), the mixed score \text{Score}(k)=\alpha\cdot|\text{PACF}(k)|+(1‑\alpha)\cdot\text{MI}(k) is calculated, and the feature window is optimized by combining forward selection and backward pruning (correlation coefficient threshold 0.8). 4. Technical effect: The fault warning accuracy is improved by 17.7% (wind power data); the feature redundancy is reduced by 35%-53%; the data processing efficiency is improved by 50%.

The present invention provides a road damage warning method based on deep learning, comprising the following steps: S101, constructing a road damage dataset, or using an existing public road damage dataset: the road damage dataset is collected by installing a camera on a car during driving on the road, the collected image data is manually annotated, and the damaged position is framed and typed using an annotation tool; S102, all images in the road damage dataset are preprocessed, and a bilateral filtering operation is performed to obtain a processed road damage dataset; the present invention can quickly extract features of road damage images by partially convolving the convolution layers in the FCB module and the FCP module, and improving the pooling layer, thereby enhancing the ability of the road damage warning model to process complex road scenes and reducing feature redundancy of road damage images. In the CSAttention module, a channel attention mechanism is introduced, the model is simple, and the running speed is fast.

The invention discloses a marine route planning method considering uncertain factors, and belongs to the field of marine traffic safety management, and the method comprises the steps: S1, carrying out the preprocessing and feature fusion of multi-source data; s2, carrying out the uncertainty modeling of the marine route, and quantifying the probability distribution of uncertainty parameters; s3, predicting a candidate route trajectory based on PCA hierarchical attention; s4, in combination with the probability distribution of the uncertainty parameters and the predicted candidate route trajectory, calculating the operability super probability of each segment of the route, and generating a probability operability index; and S5, by adopting an approximate MinSumA algorithm, outputting an optimal route by taking the integration of the minimum total navigation time and the maximum probability operability index as targets. By the adoption of the sea route planning method considering the uncertainty factors, a data-modeling-prediction-evaluation-optimization closed loop is constructed, the problems that a traditional method ignores uncertainty, is high in optimization complexity and poor in real-time performance are solved, and safety-efficiency-real-time performance collaborative optimization is achieved.

The application discloses a liver cancer prognosis evaluation model generation method and related devices, and relates to the technical field of data processing. The method obtains medical multi-modal data of a liver cancer patient, the medical multi-modal data including macroscopic magnetic resonance imaging data and microscopic pathological whole slice image data; based on a graph neural network, the medical multi-modal data is converted into multi-scale graph structure data and is subjected to cross-scale topological alignment through a GroMoVe optimal transport algorithm to obtain topological fusion features; the topological fusion features are input into a causal intervention module constructed based on a structural causal model, a counterfactual generation is performed to eliminate confounding factors, and causal invariant features are extracted; a continuous time evolution component is trained based on the causal invariant features, the continuous time evolution component represents time dynamic changes of a tumor recurrence risk based on a neural ordinary differential equation, and a liver cancer prognosis evaluation model is obtained. The application improves liver cancer recurrence prediction accuracy, enhances model generalization robustness and interpretability.

The present invention discloses a standard text classification method based on machine learning, which belongs to the field of data recognition technology, including adopting the BERT algorithm based on constrained attention to vectorize the standard text; adopting the generative adversarial network algorithm based on semantic focus to generate samples, thereby realizing the expansion of the standard text data after vectorization; and adopting the neural network algorithm based on perturbation fuzzy fusion to extract features and classify the standard text data after vectorization. The present invention uses an interactive attention mechanism and a dynamic context perception mechanism, so that the model can not only capture the semantic association within the text, but also better understand the relationship between the standard text and the classification task instructions, significantly improving the classification model's ability to understand the context. A generative adversarial network based on semantic focus is used to generate additional sample data, and the semantic quality and consistency of the generated data are ensured through a semantic refocusing mechanism, solving the problem of insufficient samples in the standard text classification task.

The present invention discloses a network intrusion detection method based on traffic classification, comprising the following steps: 1. Processing network traffic data to obtain a total sample set; 2. Establishing a network intrusion detection model DCADenseIncept based on a deep self-attention mechanism and a dense Inception structure to extract primary features from network traffic data and generate a feature representation X. L ; 3. After DCADenseIncept, a network intrusion detection model CeIT based on the convolution-enhanced Transformer structure is attached. It combines the advantages of convolution feature extraction, spatial local enhancement and Transformer to establish long-range dependencies, and provides a new model for X L Extract high-level features and generate feature representation X H ; 4. Represent the feature X H The classifier is used to perform classification detection and calculate the overall loss, and the Adam optimization algorithm is used to automatically update the model parameters to obtain the final network intrusion detection model. The model in the present invention can enhance the processing method and utilization efficiency of features, has excellent expression ability and network intrusion detection effect, and can assist the system in resisting network attacks.

The present invention relates to a detection method for defects in a three-layer conductive structure. The method uses Fourier transform and Hilbert transform to analyze the frequency domain and instantaneous frequency domain of a pulsed eddy current detection signal and extract features, and extracts features such as peak value and zero-crossing time in the signal time domain, so as to obtain defect information of the three-layer conductive structure better and comprehensively; and adopts a feature dimension reduction method based on AIC and Fisher to reduce feature redundancy, so that the reduced-dimensional features can better characterize the defect type, improve the recognition accuracy and efficiency of the classifier, and finally use SVM to construct a classifier, and use a particle swarm algorithm to optimize the classifier model parameters, so as to improve the training speed and accuracy. The present invention can improve the precision and recall rate of defect position and defect size recognition of the three-layer conductive structure, and can reduce the error of defect size recognition.