41results about How to "Fully learn" patented technology

The invention provides a zero-sample dynamic MRI reconstruction method based on global and local diffusion models, and the method comprises the following steps: generating an image data set based on original two-dimensional dynamic full-sampling heart magnetic resonance multi-coil data; designing a global and local time interlaced sampling strategy to construct two types of training data sets for generating a network model by adopting global and local cascade learning networks of a two-stage learning mechanism; performing prediction, correction, low-rank constraint and data consistency on data alternation by using an iterative solution algorithm for rapidly extracting prior information of a network model; and applying an iterative solution algorithm to the trained network model, testing, outputting a dynamic magnetic resonance reconstruction image, and evaluating the image quality. According to the method, the complete coding data is constructed through different time interlaced sampling strategies, the image is quickly and accurately reconstructed under the under-sampling condition in combination with the image prior information of the network model, and the quality and efficiency of the reconstructed image are remarkably improved.

The invention relates to a slope monitoring method and device based on multi-modal data, and the method comprises the steps: obtaining monitoring data of a target slope, the monitoring data comprising interference radar data and multi-spectral data at different moments obtained by monitoring the target slope; inputting the monitoring data into the trained preset model to obtain a monitoring result; wherein the monitoring result represents the change condition of the target slope at different moments, the training sample is used for training the preset model to obtain the trained preset model, and the training sample is obtained based on the digital elevation data, the interference radar data and the multispectral data. As the training sample comprises the digital elevation data, the interference radar data and the multispectral data, the model can learn the change information of the side slope in the three-dimensional space in the time dimension, and the interference radar data and the multispectral data in the side slope can be more fully learned. Therefore, the obtained monitoring result can reflect the slope change condition more accurately, and then the slope monitoring efficiency is improved.

The embodiment of the invention relates to a processing method and device for carrying out structure optimization on electrolyte formula components. The method comprises the following steps: constructing a structure optimization model; constructing a first data set and a second data set; performing pre-training based on the first data set, and performing tuning based on the second data set; after the optimization is finished, taking an electrolyte formula input by a user as a current electrolyte formula, performing corresponding three-dimensional molecular structure modeling according to each component molecular sequence of the current electrolyte formula, and performing three-dimensional structure optimization on each component modeling structure based on a structure optimization model; forming a corresponding formula component structure optimization record by each component molecular sequence, the corresponding component type and the component optimization structure; and forming a corresponding formula structure optimization report by all the obtained formula component structure optimization records, and feeding back the report to a user. According to the method, the influence of human factors on the molecular structure modeling quality can be reduced, and the modeling quality and the modeling efficiency are improved.

The invention discloses a step-type landslide displacement prediction method based on a VMD-Mama algorithm. The step-type landslide displacement prediction method comprises the steps of constructing a data set in the same region, performing variational mode decomposition, performing variational mode decomposition, performing periodic term displacement component prediction and performing prediction component superposition and prediction precision evaluation. According to the invention, the unit prediction model and the multivariate prediction model are respectively constructed for different displacement components based on the Mama model, and the prediction precision of the accumulated displacement of the landslide is improved.

The invention discloses an unmanned aerial vehicle low-altitude visual angle target detection method and device based on deep learning, and belongs to the technical field of computer vision. The method comprises the following steps: performing data enhancement on a collected target detection object image data set, and dividing the enhanced target detection object image data set into a training set, a verification set and a test set; an attention mechanism module is added in a backbone network of the YOLOV7-OBB model; the YOLOV7-OBB model is trained based on the training set, the YOLOV7-OBB model in the training process is evaluated based on the verification set, and the trained YOLOV7-OBB model is obtained; and inputting the test set into the trained YOLOV7-OBB model, and carrying out target detection to obtain a detection result. By enhancing a target detection object image data set and adding an attention mechanism module, the YOLOV7-OBB model fully learns key feature information of a target, and the detection precision and robustness are remarkably improved.

The invention provides a flight ground guarantee data filling model construction method, which comprises the steps of constructing a data filling model, comprising a time sequence position coding module, a topological feature coding module, a mask matrix generation module, a time sequence matrix generation module, a first splicing module, a feature fusion module, a pre-filling matrix generation module, a second splicing module, a filling module, a first updating module, a third splicing module, a reconstruction module, a weighted fusion module and a second updating module. Obtaining an initial time sequence characteristic matrix set R based on the original data in the set time period, and dividing the R into a training set, a verification set and a test set according to a preset proportion; training the initial data filling model by using the training set to obtain a trained data filling model; and verifying the trained data filling model by using the verification set, and testing the final data filling model by using the test set. According to the invention, the data filling accuracy can be improved.

The invention relates to a generation type span-based named entity recognition method in big data, which comprises the following steps of: inputting a text sentence into a trained named entity recognition model, and outputting an entity and an entity category in the text sentence; wherein the named entity recognition model comprises a feature extraction module, a candidate span generation module and a candidate span classification module; the feature extraction module is used for performing feature extraction on input text sentences to obtain feature vectors of each word in the text sentences; the candidate span generation module is used for taking the word meeting a preset condition in the text sentence as a span seed according to the feature vector of each word, and generating a candidate span entity according to the span seed; the candidate span classification module is used for classifying the candidate span entities according to a preset template to obtain the categories of the entities, flat entities and nested entities can be recognized at the same time, the calculation amount is reduced, and the recognition efficiency and accuracy are improved.

A wind turbine generator variable pitch bearing fault diagnosis method based on signal coding specifically comprises the steps that original vibration signal data of a variable pitch bearing are acquired through a wind turbine generator state monitoring system arranged in a wind power plant; bearing vibration signals are respectively coded into a time-frequency diagram and a position matrix diagram by adopting continuous wavelet transformation and position matrix transformation, and the characteristic information of the signals is enhanced by increasing the visual dimension of data; a global feature extraction module and a local feature extraction module are respectively designed based on a global attention mechanism and a window attention mechanism, so that the model can fully learn feature information of different levels; a dual-channel parallel fusion network model is constructed according to global and local feature extraction modules to fuse global and local feature information, and the purpose of suppressing strong random noise interference is achieved. According to the method, the vibration signal characteristics of the variable pitch bearing in different states can be fully captured, and the problems of poor diagnosis performance and low precision of the variable pitch bearing caused by strong random noise interference can be effectively solved.

The invention discloses a multi-modal false information detection method, system and device and a storage medium, and belongs to the technical field of false information detection.The method comprises the steps that a text of multi-modal information is input into a text representation network to obtain text representation, and an image of the multi-modal information is input into an image representation network to obtain image representation; and inputting the text representation and the image representation into a shared classifier to obtain a text prediction information label, an image prediction information label, a text weight and an image weight, and weighting the text prediction information label and the image prediction information label based on the text weight and the image weight to obtain a prediction information label. According to the method, the text and image modal information is combined, and an adaptive weight distribution mechanism is adopted, so that efficient and accurate multi-modal information processing is realized, the prediction performance is remarkably improved, and the application range is widened.

The invention discloses a bolt missing detection method based on cross-media and knowledge reasoning. The method comprises the following steps: constructing a bolt-fitting image-text pair data set and a bolt knowledge graph; a cross-media bolt knowledge fusion model is created, and the cross-media bolt knowledge fusion model generates overall text description of the image by fusing global information extracted by the fine-tuned MiniGPT-4 and local information captured by the YOLOv8; and creating a big language reasoning model based on the knowledge graph, guiding the big language model to perform knowledge reasoning on output information of the cross-media bolt knowledge fusion model by using the bolt knowledge graph through the big language reasoning model based on the knowledge graph, and obtaining an interpretable bolt missing detection result. According to the method, the collected bolt image can be detected, the bolt missing detection precision is improved, the bolt missing detection result has interpretability, and compared with other advanced missing bolt detection methods, the AP value and the mAP value are optimal.

The invention belongs to the field of multi-label image classification, and particularly discloses a neural collapse-based long-tail multi-label image classification method, which guides learning of different types of features by using label embedding of an equiangular tight frame, so that the different types of features meet the property of neural collapse, the difference between different types is maximized, and the neural collapse-based long-tail multi-label image classification efficiency is improved. The influence of the head class on the tail class in the same sample is reduced, and learning of tail class features is fully guaranteed. Besides, a comparative learning method is also adopted, so that each class feature is clustered and collapsed to a class prototype, and the model learns more robust features. And finally, features corresponding to different classes of samples extracted by the model are combined and input into a binary classifier meeting an equiangular tight frame, so that the sample features are fully utilized, the influence of an unbalanced data set on the training of the model is avoided, the generalization ability of the model is enhanced, and the training efficiency of the model is improved. And the prediction precision of multi-mark image classification is improved.

The invention discloses a facial acne grading method based on collaborative mapping multi-scale self-distillation, is applied to the field of computers, and aims to solve the problem of low accuracy of acne severity grading in the prior art. According to the method, the convolutional neural network model is trained by adopting a self-distillation architecture, and a novel distillation module optimization training process is provided in combination with the characteristics of acnes, so that the model can notice the uniqueness of acne images, and the severity grading precision of the model for facial acnes is improved. The method can automatically identify the acne features and grade the severity of the acne features, assists doctors in diagnosis and treatment, and accelerates the healing period of the acne.

The invention provides a dialect recognition method, device, equipment and program product, and the method comprises the steps: inputting dialect voice into a dialect voice recognition model, and obtaining a dialect recognition result; wherein the dialect speech recognition model is obtained by performing dialect speech recognition training based on a dialect speech sample and performing dialect text prediction training based on a dialect text pronunciation sample, and the dialect text pronunciation sample comprises mandarin pronunciation corresponding to a dialect text. According to the scheme, the dialect speech recognition model is trained based on the new model training scheme and is used for dialect speech recognition, and the dialect speech recognition effect can be improved.

The invention provides a marine diesel engine parameter datum line prediction method based on a random forest algorithm. The marine diesel engine parameter datum line prediction method based on the random forest algorithm comprises the following steps that a Waxtin 9L34DF dual-fuel supercharged engine is modeled and calibrated according to operation data of a real ship, and datum line data of an engine model under different working conditions are obtained in AVL-BOOST software; carrying out normalization processing on the datum line data; dividing the normalized data into a training set and a test set according to a certain proportion; inputting the training set data into the random forest prediction model, training the model, and stopping training and storing the training set data into the random forest prediction model after the random forest prediction model reaches a set number of iterations; and inputting the data of the test set into the random forest prediction model stored through training, classifying the data of the test set, and performing datum line prediction under a dynamic working condition to obtain a corresponding prediction result. According to the invention, the real environment in the cabin can be simulated by adopting a noise increasing mode according to the data collected in the model.

The invention relates to the technical field of image recognition, in particular to a sleep post recognition method and system. The method comprises the following steps: collecting a sleep post video sample; extracting a sleep post picture sample from the sleep post video sample; labeling a label of each sleep post picture sample; determining a target detection area in the sleep post picture sample; initial feature extraction is carried out on the target detection area; determining a spatial attention feature of each initial feature based on a spatial attention mechanism; fusing the initial features and the spatial attention features to obtain new features; constructing a training sample by using the new features and the labels, and training by using the training sample to obtain a sleep post identification model; acquiring a to-be-detected video / picture in real time; whether the person in the to-be-detected video / picture is in the sleep state or not is judged by using the sleep recognition model. By adopting the method, the accuracy and accuracy of sleep recognition can be improved.

The invention discloses a cable-stayed bridge structure displacement effect analysis method and system, and the method comprises the following steps: S1, data collection and processing: collecting monitoring data related to the displacement response of a cable-stayed bridge, and carrying out the preprocessing of the data, and obtaining a data set; s2, data set expansion: performing spectral analysis on time series data of the cable-stayed bridge by taking a timestamp as an index, performing feature engineering and time-frequency feature extraction to realize expansion of an existing feature number, expanding the expansion data into a data set, and randomly dividing the data set into a training set and a test set; s3, constructing a model to obtain a structure displacement prediction model; and S4, model testing. According to the method, the learnable normalization and anti-normalization technologies are combined, dynamic circulation closing of processes such as data, models, model explanation and explanatory feedback is achieved, the generalization performance of the prediction model is improved, and practical application of a cable-stayed bridge structure displacement effect analysis algorithm model is facilitated.

The invention discloses a multi-layer semantic perception, distillation and semi-supervised cooperative training target detection method, and the method specifically comprises the following steps: constructing a data set: extracting a first part of images from image data, marking the first part of images to construct a supervised target detection data set, and taking the remaining images as an unmarked image data set, the data volume of the unlabeled image data set is greater than that of the supervised target detection data set; teacher model optimization: performing supervision training on the teacher model on the supervised target detection data set; constructing a teacher model, and executing self-distillation training on the teacher model; pseudo labels are generated for the unlabeled images through a teacher model, and adaptive screening is carried out based on confidence distribution; mapping a pseudo label to a strong enhanced sample through enhanced geometric transformation, carrying out semi-supervised training by using the strong enhanced sample and the pseudo label, and introducing a feature layer distillation constraint at the same time; optimizing a student model; and outputting the target detection model obtained through training.

The invention provides a network intrusion detection method and a network intrusion detection device based on deep learning, and the method comprises the steps: sequentially inputting the input data in multiple groups of current training data into a convolutional neural network layer, a gating time sequence convolutional network layer and an attention mechanism layer of a network intrusion detection model, so as to obtain sample target feature data; outputting a corresponding sample traffic data type through the full connection layer; performing model verification according to all the sample flow data types to obtain model precision; under the condition that the model precision does not meet the precision requirement, all parameters in the network intrusion detection model are adjusted until the model precision meets the precision requirement, and a target network intrusion detection model is obtained; and inputting the network traffic data into the target network intrusion detection model, and outputting to obtain a real-time traffic data type. The problem of low network intrusion detection precision caused by the fact that time features and space features are not combined to extract important feature information during network intrusion detection in the prior art is solved.

This disclosure relates to a training method for a multi-label recognition model for speech recognition and a smart home device. The multi-label recognition model includes a shared feature extraction network and multiple branch recognition networks. The shared feature extraction network outputs shared features based on audio data, and the multiple branch recognition networks output multiple recognition labels based on the shared features. The training method includes: acquiring training data, which includes audio data and the ground truth value of at least one of multiple recognition labels associated with the audio data; using the multi-label recognition model to output a predicted value for each of the multiple recognition labels based on the audio data; calculating a branch loss function value between the ground truth value and the predicted value for each of the at least one recognition label; calculating a total loss function value based on the branch loss function value; and updating the parameters of the multi-label recognition model based on the total loss function value.

The present application relates to the technical field of aerospace artificial intelligence, and provides a space debris collision risk probability prediction method based on deep learning, which takes ESA Kelvins real rendezvous data as a benchmark, and uses Kessler to generate a synthetic CDM sequence to expand the training sample to improve the coverage of high risk and complex situations; at the same time, a plurality of update records and their associated related events are expressed as a dynamic heterogeneous graph, a dynamic graph neural network is used to learn the law of risk evolution over time and the similarity between events can be optionally used for knowledge transfer, so that the continuous regression prediction result of the final collision risk probability of the rendezvous event can still be stably output under the condition of missing and noise, and the prediction is quickly refreshed when new CDM arrives through the local graph incremental update mechanism, which significantly improves the accuracy, robustness and online availability of risk prediction.

The invention discloses a hyperspectral image super-resolution method based on a double-domain gating convolution attention mechanism, which can improve the attention of high-frequency information of a hyperspectral image and improve the super-resolution effect after feature fusion. The method comprises a depth feed-forward convolution module, a non-bureau gate control convolution module, a dual-domain gate control attention module and a reconstruction module. According to the method, firstly, a spatial high-frequency information region in an image is filtered by adopting non-local department controlled convolution (NGC), and meanwhile, attention to a global context is kept through a self-adaptive high-frequency component attention mechanism; under the guidance of spatial domain-spectral domain correlation, a hyperspectral image is reconstructed without influencing details, and then spatial high frequency and texture information of the hyperspectral image is reserved through depth feed-forward convolution (DFC) and double-domain convolution attention. Experimental results prove that the algorithm is superior to other common remote sensing image super-resolution algorithms in subjective visual effects and objective evaluation indexes.

The invention discloses a risk identification method and device, equipment and a storage medium. In the scheme, transaction log information to be identified is acquired; and generating risk prompt information based on the transaction log information and the corresponding business scene. And splicing the transaction log information with the risk prompt information to obtain to-be-identified risk information. And inputting the to-be-identified risk information into a target risk identification model constructed based on a BERT model for risk identification to obtain a risk identification result. According to the scheme, the risk information is enhanced by generating the specific risk prompt information for different business scenes, risk identification is performed on the enhanced information through the risk identification model constructed based on the BERT model, the risk information is fully captured, and the risk identification accuracy is integrally improved.

The invention discloses a train transmission system fault diagnosis method based on parameter alignment, and the method comprises the steps: S1, constructing a personalized federated learning system which comprises a server and a plurality of clients; s2, the server issues the initialization model to each client; s3, carrying out local training by the clients by utilizing respective data, and calculating an aggregation weight; s4, the server generates a new global model and issues the new global model to each client; s5, executing a cross-round parameter semantic alignment strategy by the client, and updating the local model; repeating the steps S3 to S5 until a preset maximum training round is reached; and S6, inputting the state data of the train transmission system into the client local model after the maximum training round is ended, and outputting a fault diagnosis result of the train transmission system. According to the invention, on the premise of ensuring data privacy of each client, efficient cooperative training of the global model and the local model is realized, and the overall accuracy of fault diagnosis of the train transmission system is improved.

The invention relates to the technical field of signal processing, and discloses a simultaneous same-frequency full-duplex self-interference elimination method and system, and the method comprises the steps: carrying out the data preprocessing of a generated self-interference signal, generating a self-interference signal containing a nonlinear factor, and constructing a training data set according to the self-interference signal; establishing a plurality of CNN-LSTM attention models for extracting nonlinear features and time sequence features of the self-interference signals; an auxiliary sub-module is introduced in the training stage of the model, so that the model has internal knowledge migration, and knowledge distillation of a feature level is realized; training the model, and performing parameter optimization by adopting a loss function which considers the amplitude and the phase of the complex signal at the same time; based on the trained model, linear and nonlinear two-stage self-interference elimination is carried out on the received signal, self-interference suppression in a full duplex system is realized, and the problems that the existing full duplex nonlinear self-interference elimination method based on deep learning is insufficient in feature extraction capability, limited in supervision information and low in training speed are effectively solved through the method.

The invention provides a CNN, BiLSTM and Attention-based photovoltaic power generation prediction method and system. The method comprises the following steps: carrying out normalization processing on a photovoltaic power generation data sequence; preliminarily extracting time features of the photovoltaic power generation data sequence through a one-dimensional convolutional neural network (CNN) and performing dimension reduction; time features are further extracted through a bidirectional long and short time memory network BiLSTM, a forward layer carries out forward calculation to obtain the output of each step length to a front hidden layer, a backward layer carries out backward calculation to obtain the output of each step length to a rear hidden layer, and finally output feature sequence vectors are obtained through combination; key features are extracted through an attention mechanism Attention in combination with the feature sequence vectors and are combined according to the importance degree, attention weights are distributed to the feature sequence vectors, weighted summation is carried out, and complete features of the photovoltaic power generation sequence are obtained; and constructing a photovoltaic power generation prediction model and predicting the photovoltaic power generation power in combination with the complete features. According to the method, the photovoltaic power generation data sequence can be fully learned, and the prediction efficiency and accuracy are greatly improved.

The invention discloses a short-term offshore wind power prediction method, and the method comprises the steps: S1, collecting wind power time sequence data, and carrying out the preprocessing of the wind power time sequence data; s2, realizing fine division of the wind power plant group based on multi-source fusion spatial-temporal clustering analysis, and obtaining spatial-temporal correlation among the wind power plants; s3, constructing an offshore wind power prediction model based on a bidirectional space-time convolution module, a space-time attention module and a residual feature fusion module, and performing preliminary prediction on the offshore wind power; s4, performing maximum overlapping double-domain cooperative decomposition on the prediction error of the preliminary prediction of the offshore wind power, and extracting error information of different frequency bands; s5, introducing an attention mechanism to dynamically distribute weights, focusing key information to drive model feedback learning, improving offshore wind power prediction precision, and outputting a prediction result; and S6, performing comparative analysis, error analysis and economical efficiency analysis on the output prediction result. The problem that an existing offshore wind power prediction model is poor in prediction precision is solved.

The invention provides a manufacturing service recommendation method based on a knowledge graph and a graph convolutional network, which fully excavates information contained in manufacturing services and tasks in a manufacturing service-task information knowledge graph, and realizes feature representation of the manufacturing services and the tasks through the graph convolutional network. In the process of aggregating manufacturing service neighbor information and task neighbor information, neighbors are selected according to degree centrality of nodes, neighbor information is selectively aggregated according to relation scores and self-attention scores of manufacturing service neighbor nodes, noise is reduced, manufacturing service representation is generated, and task neighbor information is aggregated by using a self-attention mechanism. The problem that the recommendation result has limitation due to the fact that modeling representation at the other end is insufficient is effectively solved, and the accuracy of the recommendation result is improved.

A panoramic interactive tooth segmentation method based on feature enhancement comprises the following steps: reconstructing a two-dimensional panoramic image on the basis of a three-dimensional image of an original cone beam CT, calculating a three-dimensional initial seed point and a single-tooth region of interest, and generating final three-dimensional foreground and background seed points after semantic discrimination and adjustment; generating corresponding foreground and background distance maps, inputting the foreground and background distance maps and a single-tooth cone beam CT image into a Transform model at the same time, calculating and generating intra-layer and inter-layer features, fusing the intra-layer and inter-layer features to generate foreground and background enhanced distance maps, performing image segmentation by using 3DUNet, and outputting a single-tooth initial segmentation result; and evaluating the initial segmentation result, performing secondary interaction on the slices of which the scores are lower than a threshold value by the user to generate secondary seed points, and finally optimizing the initial segmentation result according to the secondary seed points. The method is implemented on the two-dimensional panorama, so that the interaction robustness is improved; the foreground enhanced distance map and the background enhanced distance map serve as two independent sources, and compared with a traditional mode that only foreground seeds serve as interaction sources, the difference characteristics between the background and the foreground can be fully learned.

The invention belongs to the field of FTU fault identification, and particularly relates to a zero sequence voltage abrupt change characteristic grounding fault identification method and system for FTUs, and the method comprises the steps: firstly obtaining a power distribution network bus zero sequence voltage discrete sequence, calculating a continuous sampling point difference absolute value sequence, and constructing an abrupt change characteristic quantity sequence through sliding time window integration; the time window features are input into a pre-trained capsule neural network, and the primary capsule layer encodes the time window features into primary capsule vectors; and updating the connection weight between the primary capsule and the high-level capsule to be converged through iterative computation, and judging a fault according to the converged high-level capsule vector: judging the fault and outputting a starting signal when the model length of the capsule in a grounding fault state exceeds a first threshold value and the included angle between the capsule and the fault feature base vector is smaller than a second threshold value, otherwise, judging that the system is normal. The FTU identification precision and efficiency can be improved.

The invention relates to the technical field of computer vision and natural language processing, in particular to a video clip sentence positioning algorithm based on progressive interaction and multi-modal alignment. The method comprises the following steps of: firstly, sorting all query sentences related to a video according to the sequence of video clips corresponding to the query sentences in the video, and then splicing the query sentences with video features and feature dimensions; extracting intra-modal information of video and query sentence features and information between the two modals through a multi-modal alignment module, generating a plurality of candidate segments according to the video features interacted with the query sentence, and learning the relationship between the candidate segments through a grouping candidate segment interaction module; then reducing the difference between corresponding candidate segment features and query sentence feature pairs through metric learning, finally calculating matching scores of all candidate segment features and a single query sentence feature, and taking the feature with the high score as a prediction result. According to the invention, the video clip can be accurately positioned.