13459 results about "Convolution" patented technology

The invention provides a road bridge crack detection method and system, and the method comprises the steps: collecting a bridge surface multi-view image, and constructing a training data set containing crack feature labeling through quality screening and standardized labeling; preprocessing the image by using a multi-scale feature fused deep convolutional neural network and carrying out semantic segmentation, initially identifying a suspected crack region and generating a segmentation mask; and constructing a BeNNS proxy model based on the mask, and establishing a mapping relationship between the detection result and the bridge structure topology, the stress flow field and the service function chain so as to evaluate the result reliability. And inputting an evaluation result into a hybrid evaluation mechanism, performing online real-time detection and offline batch verification to optimize precision, and outputting a verified crack region. Finally, morphological analysis is conducted on the area, geometric parameters and danger levels of cracks are extracted and integrated to a bridge health monitoring system, a crack evolution tracking algorithm and an early warning mechanism are established, and dynamic tracking early warning is achieved. The problem of low detection precision in a complex environment can be solved.

The invention provides a self-adaptive network topology dynamic reconstruction method and system based on deep reinforcement learning, and relates to the technical field of deep reinforcement learning, and the method comprises the steps: obtaining the topology state information, service flow distribution information and historical reconstruction records of a current network; extracting topological correlation characteristics among nodes through graph convolution operation, and generating fusion state representation in combination with service flow information; inputting the fusion state representation into a deep reinforcement learning model to identify bottleneck nodes and redundant links, and outputting a reconstruction action candidate set; searching and evaluating the long-term cumulative income of the candidate actions through a Monte Carlo tree, and screening an optimal reconstruction action sequence; a graph coloring algorithm is utilized to allocate time slots and process resource conflicts, and a resource-feasible topology adjustment scheme is generated; and extracting a network evolution rule through tensor decomposition, and constructing a topological optimization association mapping graph. According to the method, the network bottleneck can be intelligently identified, the network topology structure is dynamically optimized, and the network performance and the resource utilization rate are effectively improved.

The invention provides an engineering construction defect automatic detection and classification method based on deep learning, and the method comprises the steps: obtaining a welding seam surface image through the shooting of an unmanned plane, and carrying out the denoising and illumination normalization processing of the welding seam surface image, and obtaining a standardized image; welding seam surface texture features are extracted from the standardized image, a convolutional neural network is adopted to analyze the spatial distribution characteristics of textures, and vectorization processing is carried out to obtain texture feature vectors; segmenting a weld surface corresponding to abnormal region distribution by adopting a region growing algorithm, and analyzing pore and weld discontinuity in combination with the texture feature vector to obtain a defect candidate region; performing threshold division on the sizes and the numbers of the defects according to the defect types and the feature vectors of the candidate regions to obtain a severity grading result of each type of defects; and severity features are extracted from a grading result, and a Bayesian network is adopted to fuse texture feature vectors and defect type labels to obtain a welding quality evaluation score.

The invention discloses an image processing method and device and a storage medium. The method comprises the steps of obtaining a to-be-simulated 3D convolution model and training data; decomposing the 3D convolution model into cascading of a 3D space convolution model and a 3D time convolution model to obtain a pseudo 3D cascading convolution model; training a pseudo 3D cascade convolution modelby using the training data, and obtaining parameters of a 3D spatial convolution model and a 3D time convolution model; converting the 3D space convolution model and the 3D time convolution model intoa 2D space convolution model and a 2D time convolution model; setting a feature rearrangement rule for the 2D spatial convolution model and the 2D time convolution model; mapping model parameters ofthe 3D spatial convolution model and the 3D time convolution model into parameters of a 2D spatial convolution model and a 2D time convolution model to obtain a 2D cascaded convolution model; and performing convolution operation on the image by using the 2D spatial convolution model and the 2D time convolution model. By means of the mode, image processing conducted through 3D convolution operationcan be achieved through the 2D convolution model.

The invention relates to the technical field of aircraft structure detection, and discloses an aircraft structure crack intelligent identification method based on deep learning. The method comprises the following steps: acquiring original vibration response signals and electromagnetic field distribution data on the surface and inside of an aircraft structure in parallel through a multi-source sensor network; synchronously processing the data by using a multi-scale convolutional neural network, and extracting time-frequency domain abnormal fluctuation features and space magnetic field distortion features; constructing a cross-modal correlation model, analyzing a topological dependency relationship of the two types of features through a graph attention mechanism, and generating a fused damage sensitive feature vector; inputting the vector into a pre-trained deep belief network to obtain a probability distribution mapping relation for different crack types; and according to the mapping relation, carrying out adaptive weighted fusion on original multi-sensor data, inhibiting environmental noise and structural background interference, and separating and reconstructing an accurate three-dimensional morphology map of the target crack. According to the method, multi-source data information can be effectively fused to improve the accuracy of aircraft structure crack identification.

Deep learning tools such as convolutional neural networks (CNNs) and transformers have spurred great advancements in computational biology. However, existing methods are constrained architecturally in context length, computational complexity, and model size. This application introduces a sub-quadratic architecture for modeling, which combines projected gated convolutions and structured state spaces to achieve local and global context with, for example, single-nucleotide resolution. These models outperform CNN-, GPT-, BERT-, and long convolution-based models in many tested genomics tasks without pre-training and with 4×-781× fewer parameters. In the proteomics domain, these models similarly outperform pretrained attention-based models, including ESM-1B and TAPE-BERT, on remote homology prediction without pre-training and while using 3,308×-23,636× fewer parameters.

The invention provides an intelligent drilling speed prediction method based on physical feature guidance and multi-source information fusion, and relates to the technical field of intelligent drilling speed prediction, and the method specifically comprises the following steps: collecting multi-source heterogeneous data from a drilling real-time database, a logging system, a logging system and a geological database; constructing a dual-channel deep learning prediction model, wherein the dual-channel deep learning prediction model comprises a dual-channel convolution feature extraction module, a feature fusion module, a time sequence fusion module, a time sequence modeling module and a full connection layer which are connected in sequence; obtaining a predicted drilling speed by using a dual-channel deep learning prediction model; a joint loss function is constructed by considering a data driving error and a physical constraint error, an error is calculated according to the joint loss function, and network parameters are updated through back propagation; carrying out loop iteration training until convergence; and the trained dual-channel deep learning prediction model is used for drilling speed prediction. According to the technical scheme, the problems that in the prior art, a mechanism model is insufficient in precision, and a data driving model is poor in reliability are solved.

The present invention relates to the technical field of deep learning and computer vision. Disclosed is an unmanned aerial vehicle image-based small object detection method for target areas. The present invention crops images of obvious small objects in certain target areas, and annotates the small objects of different categories to form a raw training and testing dataset, so as to ensure the accuracy of data required in the early stage of the algorithm and further ensure the scientificity of the algorithm; uses the computing capability of an improved YOLOv7 detection model to collect image features of different degrees in the dataset, the improved YOLOv7 detection model using YOLOv7 as a basic model and adding to a neck network an MS-CET module, which is constituted by an improved self-attention mechanism and convolution module SPPCSP, and a BHC-FB module, which is constituted by bidirectional mixed convolution modules NConv and RPConv connected in parallel; and finally fuses different feature layers as a final judgment basis of an unmanned aerial vehicle for small object detection in the target areas, to further check the accuracy of the algorithm and criteria for dataset selection, thereby improving recognition accuracy.

The invention relates to the technical field of computer vision, and discloses a real-time video analysis method based on deep learning. The method comprises the following steps: acquiring a real-time video stream through image acquisition equipment, and performing frame segmentation processing to generate a continuous video frame sequence; and extracting features of the video frame sequence by using a pre-trained convolutional neural network to obtain a multi-dimensional feature vector, inputting the multi-dimensional feature vector into the time sequence analysis model to calculate dynamic relevance, and outputting an inter-frame movement track and object behavior features. And constructing a scene understanding map containing a spatial position and a time evolution relationship according to the above-mentioned data, and carrying out abnormal event detection and generating event marking data based on the map. And performing semantic analysis on the event marking data, determining an abnormal event type and a confidence score, triggering a real-time alarm signal according to a result, and updating a historical event database. In the analysis process, the resource occupancy rate of the system is continuously monitored, the calculation precision is dynamically adjusted, a degradation processing mechanism is started when a preset threshold value is exceeded, and key area analysis is preferentially guaranteed.

The invention discloses a disaster intelligent Internet of Things multivariable time sequence anomaly detection method and device, and relates to the technical field of Internet of Things anomaly detection, and the method comprises the steps: S1, constructing an initial anomaly detection model; s2, acquiring a training data set; s3, performing optimization training on the initial anomaly detection model by using the training data set to obtain an optimized anomaly detection model; s4, acquiring real-time monitoring data; s5, analyzing the real-time monitoring data by using the optimized anomaly detection model to obtain a detection result; the dynamic gated expansion convolutional network DGDC solves the problems of rigid structure and parameter explosion of a traditional TCN. Dynamic expansion rate scheduling enables a receptive field to expand in an exponential level along with the number of layers, and second-level burst and week-level periodic characteristics can be captured at the same time; the parameter quantity is reduced by 60%-70% through depth separable convolution, and the efficiency and precision of local feature extraction are both superior to those of an existing convolution module by combining the suppression effect of a gated linear unit GLU on noise features.

The invention provides a resource scheduling control method and system for a big data server, and the method comprises the steps: constructing a multi-dimensional resource portrait module, collecting the CPU, memory, network, storage I / O load and task queue length of each node in real time, and predicting a resource demand trend through a time sequence algorithm; extracting characteristics such as calculation intensity, data dependence, memory requirements, network transmission quantity and the like; adjusting the weight coefficients of the resource utilization rate, the task completion time and the energy consumption efficiency according to the system load and the historical effect; establishing a bipartite graph model by taking a resource trend as a node feature and a task vector as an edge feature, and calculating a matching score through graph convolution and a multi-objective optimization function; the scheduling scheme is synchronized by adopting a consistency algorithm; automatic rollback and reallocation are carried out when resources are detected to be insufficient; and optimizing a weight coefficient and a network parameter through reinforcement learning. Through the method, the system resource utilization rate can be improved, the task execution efficiency is improved, the overall scheduling effect stability is improved, and the system fault recovery time is shortened.

The invention discloses an end-to-end automatic driving method based on dynamic multi-modal fusion in a complex scene, and belongs to the technical field of automatic driving. In order to solve the problems of sensor perception deficiency, cross-modal feature mismatching, unstable trajectory planning and the like easily occurring in night, low-illumination and complex dynamic environments in the existing end-to-end automatic driving method, texture details of a camera mode and geometric structure features of a laser radar mode are respectively enhanced through a double-flow feature refining mechanism; the characteristic difference between different modes is relieved; an information-driven dynamic fusion strategy is designed, the fusion weight is adaptively adjusted according to scene factors such as environment illumination and obstacle density, and the scene sensitivity and discrimination ability of the model are improved; asymmetric convolution and a low-rank-sparse decoupling technology are introduced, multi-order reconstruction of key channels is carried out on the multi-modal features, and the path change modeling capability is enhanced; and in combination with time sequence dependence of waypoints, outputting a future trajectory through an autoregression decoder to realize high-precision trajectory prediction and stable decision control.

The invention relates to the technical field of mineral resource management, in particular to a mineral resource dynamic prediction and mining management system which comprises a data perception and fusion layer, a unified digital twinborn model, a dynamic prediction and decision intelligent agent and a visualization and interaction control layer. The data perception and fusion layer collects structured data such as geological exploration and mining environment and market unstructured data, and generates a unified space-time tensor through processing; the unified digital twinborn model generates a dynamic comprehensive mining area situation map containing resource reserve risk economic indicators through a three-dimensional convolutional neural network embedded with an attention mechanism; the dynamic prediction and decision-making agent predicts future reserves and geological risks, and constructs a dual-objective optimization model to generate an optimal mining path equipment scheduling and resource allocation scheme; and the visualization and interaction control layer presents the mining area state and the decision scheme in a three-dimensional manner and provides an interaction interface. According to the invention, the data utilization rate and decision scientificity are improved, the safety risk is reduced, and mine management intellectualization is promoted.

The invention discloses a hybrid expert and KAN-based cyclic attention network time sequence prediction method. The method comprises the following steps: S100, inputting time sequence data needing to be predicted; s200, constructing a graph structure by using an attention mechanism, learning basic correlation characteristics among variables of the input time sequence data through an adaptive and learnable graph convolutional network, and then performing global averaging and maximum pooling on the basic correlation characteristics along a time dimension to obtain complementary time domain statistical information, so as to provide effective time-space correlation characteristics for the follow-up process; s300, after feature learning is completed, collaborative modeling of the KAN and an attention mechanism is brought into full play, rapid and efficient time sequence modeling is carried out on data by adopting a cyclic attention network embedded based on the KAN, and a foundation is laid for subsequent time sequence prediction; and S400, establishing a hybrid KAN expert-based time sequence prediction network, and adaptively fusing differentiation prediction results by a gating mechanism. The time sequence prediction method is designed from the three aspects of feature learning, time sequence modeling and time sequence prediction.

The invention discloses an earth and rockfill dam seepage-deformation early warning method and system based on time-space combined anomaly, and belongs to the field of dam body safety data research. The method comprises the following steps: constructing a spatio-temporal topological graph based on an engineering coordinate system, integrating multi-dimensional data by nodes, and constructing a dynamic adjacency matrix according to spatial distance and seepage relevance; extracting features by using a space-time diagram convolutional network, a self-loop mechanism and cross-layer attention; and executing dual-drive early warning through standard threshold preliminary screening, multi-scale LSTM prediction and a time decay evidence theory. The system comprises a sensor network and an intelligent computing module, and the intelligent computing module has adaptive modeling and visualization functions. According to the scheme, seepage-deformation space-time correlation quantitative analysis is achieved, the hysteresis effect is captured, the threshold value is dynamically corrected, multi-source evidences are fused, the early warning timeliness and accuracy are improved, and the risk of false alarm and missing alarm is reduced.

The invention discloses a power transmission line fault diagnosis and operation and maintenance scheduling method and system based on networking learning, and relates to the technical field of power transmission line fault diagnosis operation and maintenance scheduling. Related data is extracted to construct a high-risk equipment area and a visual high-risk area thermodynamic diagram, a visual risk grading diagram is constructed in combination with electrical quantity data, and meanwhile, an intelligent recognition storage network and a fault type classification recognition model are constructed in combination with a convolutional neural network-long and short-term memory network hybrid model; the model is optimized through networking learning and an attention mechanism, maintenance teams and resources are autonomously allocated in combination with an operation and maintenance management system, then autonomous optimization and closed-loop operation are achieved, full-process coverage of fault sensing, intelligent decision making and efficient response is achieved, the response time after a line fault occurs is remarkably shortened, and the maintenance efficiency is improved. And the fault handling and operation maintenance capabilities of the power grid system are comprehensively enhanced.

The invention discloses a computer memory bank fault prediction method and system based on deep learning, and relates to the technical field of computer hardware fault diagnosis, and the system comprises a multi-source time sequence data collection module which is used for obtaining memory bank operation state data in real time; the dynamic feature enhancement module is based on a composite architecture of a generative adversarial network and transfer learning, comprises a fault mode generator, and generates synthetic data consistent with real fault distribution by using an LSTM network; aligning feature spaces of different hardware platforms through a maximum mean difference loss function; the multi-modal fusion deep learning model comprises a space-time convolutional network, a graph attention network and an adaptive weight adjustment mechanism; and the fault early warning analysis module is used for analyzing a fault probability predicted value, an interpretable thermodynamic diagram and a maintenance suggestion. According to the invention, passive maintenance is changed into active prevention and control, and preposition and precision of fault management are realized through dual mechanisms of long-term trend prediction and short-term risk early warning.

The invention discloses an urban water supply management data trend analysis method based on space-time analysis, and relates to the field of data processing, and the method comprises the steps: collecting data in real time through an urban water supply pipe network sensor network, building a space-time unified coordinate system, building a space-time Kriging interpolation model based on pipe network topology, and achieving the space-time alignment of multi-source data; dividing an adaptive space-time grid by using a Voronoi diagram and a sliding window mechanism, and calculating multi-dimensional features; constructing a dynamic space-time diagram by taking a grid as a node, performing multi-step prediction in combination with a space-time diagram convolution circulation network, fusing a Kriging interpolation result, and evaluating an abnormal probability and a confidence interval through a Bayesian neural network; a monitoring layer, a prediction layer and a risk layer are overlaid in a three-dimensional GIS, a dynamic thermodynamic diagram is generated, an early warning path is optimized based on a Dijkstra algorithm, and a minimum risk topology path is output. The method has the advantages that through space-time analysis and accurate prediction, the intelligence, stability and emergency response efficiency of urban water supply management are remarkably improved, and powerful support is provided for smart city construction.

The invention relates to a water quality prediction method and system based on gating residual enhancement and feature fusion, and belongs to the technical field of water environment intelligent analysis and deep learning. Taking each water quality index as a node of the graph, and constructing two complementary variable relation graph structures by utilizing a Pearson's correlation coefficient and mutual information; respectively inputting the two graph structures into a graph convolutional network, extracting deep dependency features among indexes, and splicing and fusing the deep dependency features. A multi-head attention mechanism is used as a trunk to extract global time dependence, a GRU network is introduced to extract local time sequence features, GRU output is used as an adjustable residual term to be injected into the attention trunk through a residual gating mechanism, self-adaptive enhancement of local dynamic features is achieved, and finally a self-adaptive fusion mechanism is introduced to generate comprehensive representation. According to the method, the complex dependency relationship between the water quality indexes and the time dynamic evolution process can be modeled in a collaborative manner, the response capability to key local change and sudden change events is remarkably enhanced, and the accuracy and robustness of water quality prediction are improved.

The invention discloses a hierarchical multi-agent robot cluster control method and system, and aims to solve the problems of partial observability and environment non-stability of a multi-agent system in a complex environment. According to the method, a three-layer layered reinforcement learning architecture is constructed, a high-layer strategy is responsible for global task decomposition and role allocation, a middle-layer strategy converts tactical intention into a cooperative behavior mode, and a low-layer strategy executes accurate motion control; a graph neural network is adopted for cluster modeling, global graph representation and local neighborhood features are extracted in parallel through graph convolution and an attention mechanism, and hierarchical decision making is supported; a centralized graph enhancement evaluation network is designed to be combined with an MAPPO algorithm for collaborative optimization, and dynamic adversarial training is introduced to improve strategy robustness. According to the method, effective decoupling of global planning and local control is realized, and the cluster cooperation efficiency, the strategy interpretability and the adaptive capacity in a dynamic environment are improved.

The invention discloses a weld quality intelligent diagnosis system based on deep learning, and relates to the technical field of welding quality detection, and the weld quality intelligent diagnosis system comprises an image quality evaluation module, a feature alignment module, a deviation detection module, a path reconstruction module, a prior enhancement module and a defect identification module, identifying an area of which the signal-to-noise ratio is lower than a preset threshold value, and constructing a noise interference distribution diagram; and the feature alignment module executes a deformable convolution feature alignment operation with a confidence factor adjustment mechanism based on the noise interference distribution diagram to generate an initial space mapping result. Through mechanisms such as image quality perception, robust alignment, deviation detection, self-adaptive reconstruction and prior enhancement, a closed-loop weld joint intelligent diagnosis process is constructed, false alignment errors are effectively inhibited, the multi-modal fusion stability and the defect recognition precision are improved, and the reliability and the intelligent level of the system under complex working conditions are enhanced.

The invention, which belongs to the technical field of image analysis and geological disaster early warning, discloses a high-precision topographic change monitoring and geological disaster early warning image analysis system comprising a deformation spatio-temporal feature sensing module, a geomechanics constraint optimization module, a multi-scale disaster evolution prediction module and a self-adaptive early warning decision module. Through deep coupling and closed-loop feedback between modules, dynamic matching of deformation confidence and mechanical constraint weight, physical enhancement of stress distribution and disaster prediction, and dual-path parameter optimization driven by early warning performance are realized, and the system adopts an InSAR technology and deep learning fusion to extract millimeter-level deformation. The physical embedded neural network is utilized to realize anomaly recognition under mechanical constraints, multi-scale disaster evolution is predicted through space-time convolution Transform, and the early warning accuracy rate can reach 95% or above after closed-loop iterative optimization.

The invention discloses a cluster computing power energy efficiency perception scheduling and green computing system, which relates to the technical field of computers and comprises a multi-source energy efficiency perception and data acquisition module used for acquiring power consumption, utilization rate, temperature, cooling state, PUE index and environmental data of cluster nodes. According to the invention, through the multi-modal energy efficiency fusion sensing network and the multi-scale convolution and time sequence attention fusion network, multi-source heterogeneous energy efficiency data such as current, voltage, temperature, airflow and the like of a node level can be collected and fused in real time and with high precision, noise is effectively removed, abnormity self-correction is realized, the defect of energy efficiency sensing granularity in the prior art is made up, and the energy efficiency sensing precision is improved. And reliable input is provided for subsequent scheduling decisions. A cross-scale dynamic twinborn collaborative modeling mechanism is adopted, a physical information neural network and a computational fluid mechanics model are coupled, optimization is carried out through a generative adversarial network structure, and accurate prediction of a complex energy consumption evolution curve and a cooling flow field is achieved.

The invention discloses an unmanned aerial vehicle camera attitude estimation optimization method and device, and relates to the technical field of unmanned aerial vehicles. The method comprises the following steps: acquiring data by using an unmanned aerial vehicle carrying various sensors, preprocessing, carrying out feature extraction and matching on preprocessed image data, removing mismatching points, and based on an initial point cloud, carrying out sparse reconstruction in an incremental expansion and bundle adjustment optimization stage; integrating the re-projection error term, the flight path constraint, the epipolar geometric constraint and the triangulation constraint into a target function; constructing a model for predicting the pose correction based on the deep residual convolutional neural network, and taking the initial pose of the camera and the corresponding local image as input; and adjusting the weight of each constraint in the target function based on the predicted pose correction and pose confidence, and minimizing the target function to improve the precision of unmanned aerial vehicle camera pose estimation. The problem of reconstruction scene deviation caused by inaccurate camera attitude estimation in the prior art is solved.

The invention discloses a unified framework for solving automatic driving track prediction and planning consistency based on a world model. According to the method, through cooperative work of the automatic driving domain controller and the vehicle-mounted sensing system, end-to-end joint optimization of track prediction and planning in a complex traffic scene is realized, time sequence dependence and interaction dynamics among intelligent agents are accurately captured, and the prediction capability and robustness of a model are remarkably improved. The method comprises the following specific steps: firstly, constructing a generative world model, and generating potential future state representation by utilizing a behavior conditional and backtracking expansion technology; secondly, in combination with global modeling and a local convolutional network, multi-scale features are extracted, adaptive fusion is carried out, and a multi-modal prediction trajectory is generated; then, a multi-target planning model is adopted to integrate various driving indexes, and a track with the minimum loss function is generated; finally, path planning parameters are dynamically optimized through real-time environment perception and decision feedback, and the problems of prediction uncertainty and planning consistency of the automatic driving track are effectively solved.

The invention provides a cellular network flow space-time prediction method and system based on a hybrid neural network, and belongs to the technical field of intelligent communication. The method adopts a layered deep neural network architecture, and comprises a data embedding layer, a space-time coding layer, a feature fusion layer and an output layer. The data embedding layer maps a historical traffic sequence, cross-domain external data and metadata into high-dimensional features; the space-time coding layer is used for respectively fusing one-dimensional causal convolution and a Mama neural network to extract multi-scale time features and densely connecting convolution and a multi-head attention mechanism to capture multi-scale space features through time and space modeling branches; the feature fusion layer realizes adaptive weighted fusion of spatial-temporal features, cross-domain features and metadata features by using a gating fusion mechanism; and the output layer performs linear transformation on the fusion features to generate a final prediction result. According to the method, the spatial-temporal dynamic capture of the service traffic is accurate, the prediction curve is highly fit with the true value, and the accurate prediction of the multi-service traffic of the cellular network is realized.

The invention discloses a power transmission line state simulation and prediction method based on digital twinning, and the method comprises the following steps: accessing meteorological data, electrical data, space and asset data and historical state data, carrying out the preprocessing, and generating a multi-source spatial-temporal feature sequence set; constructing a double-domain cross-coupling time convolution twinborn model, and obtaining corresponding feature representation through a quick response path and a slow response path; physical consistency state representation is generated through a physical constraint cross gating layer; outputting a prediction result set through a simulation engine; online observation data are obtained, and the model is corrected based on virtual and real residual errors; and generating a current prediction result set by using the corrected model, and converting the current prediction result set into a scheduling and operation and maintenance strategy set. According to the method, the double-domain cross-coupling time convolution twin model and a virtual-real closed-loop correction mechanism are adopted, multi-scale state simulation prediction of the power transmission line is achieved, and the method has the advantages of being high in precision, robustness and performability.

The invention discloses a knowledge graph reasoning method based on dynamic rule perception memory. The method aims at solving the problems that an existing neural combination rule learning method is insufficient in expression ability and prone to splitting global semantics and local relation modes. The method comprises the following steps: introducing a lightweight dynamic relationship memory module, and executing self-attention on all relationships in a knowledge graph to capture global semantics; meanwhile, local relation interaction features are extracted through convolution, semantic fusion is carried out through a Transform encoder with multi-head attention and relative position coding, and unified and parallel-computing high-quality combination representation is constructed for each pair of relations in parallel. And meanwhile, a closed high-quality reasoning path is generated in combination with bidirectional breadth-first search. Through global semantic and local interactive collaborative modeling, the extendibility is ensured, the understanding of global semantics is enhanced, and the reasoning accuracy is improved. Global and local relation dependence is fused, an interpretable reasoning path is generated, and reasoning accuracy, expandability and robustness are improved.

The invention provides a medical image segmentation method based on an AFMHiFormer. The method comprises the steps that firstly, a multiple data enhancement module is provided, and the data distribution diversity is improved while the enhancement stability is guaranteed; secondly, a segmentation model AFHiMFormer is constructed, and the model architecture adopts a double-branch encoder and a multi-scale decoder; thirdly, a feature enhancement module is provided to construct a dynamic complementation mechanism of semantic enhancement and boundary modeling; fourthly, a multi-scale feature fusion module is introduced, multi-scale context information is captured through parallel hole convolution with different expansion rates, and self-adaptive fusion of global and local features is achieved; and fifth, a cross-scale fusion module is designed in the multi-scale decoder, so that the deep layer branch and the shallow layer branch are efficiently fused in a multi-level feature space. According to the method, the advantages of CNN and Transform are combined, dynamic fusion of local and global features is realized by providing a new module, and a remarkable performance advantage is shown in a medical image segmentation task.

The invention discloses a river flow missing data reconstruction method, which comprises the following steps of S1, constructing a river network topological structure, and quantifying hydraulic correlation; s2, spatio-temporal feature fusion and multi-source information extraction; s3, performing multi-task cooperative flow reconstruction and confidence coefficient prediction; s4, dynamic weighting and result correction of meteorological factors; and S5, performing anomaly detection and model dynamic updating. According to the method, a river flow missing data reconstruction method is set, the steps are mutually fused and used, and topology-aware space-time diagram convolution is performed: a river network topology structure is encoded into a weighted adjacency matrix for the first time, space correlation features are extracted through a diagram convolution network, and the problem that a traditional method ignores hydraulic connection is solved; a multi-task collaborative learning mechanism: synchronously outputting a flow reconstruction value and confidence, combining topological smooth constraints, and realizing reconstruction reliability quantification while ensuring precision; and meteorological dynamic weighted correction: dynamically adjusting the node weight based on the real-time rainfall intensity, and accurately adapting to the nonlinear response of the water flow in the heavy rainfall period.