39 results about "Network composition" patented technology

The invention relates to the technical field of computers, and discloses a multi-agent reinforcement learning fault diagnosis method based on edge-center hybrid optimization, and the method comprises the following steps: S1, collecting and preprocessing equipment data; s2, layered fault diagnosis: an intelligent diagnosis agent adopts a layered reinforcement learning structure and is composed of a high-layer strategy network and a low-layer strategy network; s3, edge-center hybrid optimization: designing a center strategy optimization agent, training a high-layer strategy network and a low-layer strategy network in stages, and performing compression and distillation on the trained strategy networks by adopting a teacher-student strategy structure; executing strategy fusion and global updating based on the received strategy execution information; and S4, strategy migration. According to the method, continuous learning and strategy updating are carried out in a scene in which fault samples are extremely scarce, diagnosis knowledge sharing, strategy synchronization and cross-device migration are realized by using a multi-agent cooperation mechanism, efficient deployment and operation at an edge device end are supported, and field state change is adapted in real time.

The invention discloses an AGV scheduling strategy generation method and device based on multi-agent reinforcement learning, relates to the technical field of AGV scheduling, and solves the problem that cooperative scheduling of multiple AGVs in a large-scale, high-dynamic and flexible workshop is difficult to effectively solve in the prior art. The method comprises the following steps: constructing a data set and a multi-agent decision model consisting of a GNN network, an Actor network and a Critic network, carrying out centralized training and distributed execution on the multi-agent decision model, generating a scheduling strategy for each AGV, and introducing the GNN network to help the Actor network and the Critic network to understand a relationship between a plurality of agents in an environment and the intelligence of a to-be-selected task set. And the Actor network makes strategy judgment for a plurality of intelligent agents. The Critic network not only evaluates the individual value of the intelligent agent in each step, but also evaluates the contribution of each intelligent agent to the global value, so that the intelligent agent not only realizes the individual optimization, but also needs to achieve the global optimization.

The invention discloses a short-time traffic flow prediction method based on a traffic-air interactive hybrid convolutional network, and belongs to the technical field of traffic flow prediction. According to the method, historical traffic flow and air quality data are collected and preprocessed, a fusion feature sequence is constructed through cross-modal time-space sequence interleaving recombination, and real-time traffic flow prediction is achieved through training optimization of a mean square error loss function and an RMSprop optimizer by adopting a model composed of a convolutional interaction network and a bidirectional long-short term memory network. According to the method, the space-time coupling relation of the two kinds of explicit modeling is recombined through cross-modal space-time sequence interleaving, differentiation feature extraction and two-way dependence mining are combined, the problem that real-time interactive modeling cannot be achieved through a traditional method is solved, and robustness and adaptability in the complex environment are improved.

The invention relates to an AMT signal denoising method and device based on self-adaptive multi-level U-Net. According to the method, noise segment classification is performed on a noisy AMT signal, a corresponding position mask of the clean AMT signal is recorded, and noise positioning information is provided for training of a denoising model, so that the training effect of the denoising model is improved, and loss of non-noise segments can be avoided; the method also uses two U-Net networks to form a U-Net cascade structure, the first network performs preliminary denoising, the second network processes fragment data output by the first network, generates a noise intensity probability by using a residual error, uses the noise intensity probability as a driving signal, distributes a corresponding weight value, and outputs the driving signal to the U-Net cascade structure. According to the method and the system, the second network is dynamically controlled to selectively absorb and fuse fragment data output by the first network, an area with remarkable residual noise can be adaptively concerned, complementation and optimization of a feature level are realized, and finally, the denoising effect of the denoising model on the noisy AMT signal is improved.

The invention discloses a self-supervised monocular depth estimation method based on an enhanced multi-scale pose network, and relates to the field of computer vision. The problems that in an existing self-supervision monocular depth estimation method, the pose network structure is simple, the time sequence modeling capacity is insufficient, and geometric constraints are missing are solved. According to the method, a self-supervision joint training framework composed of a depth estimation sub-network and an enhanced pose estimation sub-network is constructed; wherein the pose estimation sub-network extracts multi-scale spatial structure features through a layer-by-layer feature fusion encoder, and adopts a context fusion decoder based on time sequence attention to model a motion dependency relationship between continuous frames; meanwhile, a self-supervised pose consistency loss function is introduced, geometric continuity of a camera track is enhanced through forward and reverse transformation consistency constraint and closed-loop geometric constraint, and collaborative optimization of depth prediction and pose estimation is realized. The method is also suitable for the application fields of automatic driving, robot perception, augmented reality and the like.

The invention belongs to the technical field of communication, and particularly relates to a router system service exception self-healing method based on cloud AI, which comprises the following steps: deploying an exception detection module at a router end, collecting system logs and state data in real time and generating a standardized exception report; uploading the report to a cloud AI server through an encrypted MQTT protocol; the cloud calls a hybrid analysis model composed of a rule matching engine, a machine learning classifier and a reinforcement learning decision network to generate a self-healing strategy instruction packet; the router end receives and executes the strategy, completes service restart, configuration rollback or hotfix loading and other operations, and verifies the self-healing effect; when communication interruption exceeds a threshold value, an embedded loopback self-healing subsystem is automatically activated, and abnormity is independently handled based on a local strategy library. According to the technical scheme, millisecond-level abnormal response and high-success-rate autonomous recovery can be achieved, the network availability and the service continuity are remarkably improved, and the disaster recovery self-healing capacity is still achieved when the cloud end is disconnected.

The application provides a video summarization method based on multi-dimensional features and fine-grained hierarchical modeling, and relates to the technical field of video processing. In practical application, the video summarization technology can facilitate large-scale video retrieval and browsing. The method comprises the following steps: firstly, frame extraction is performed on an input video to obtain a frame sequence, and a multi-dimensional feature extraction network composed of a 2D network and a 3D network is used to extract multi-dimensional features; then, hierarchical temporal modeling is performed to complete the modeling process of the temporal dependence of the entire video sequence; finally, a regression network is used to obtain the importance score of each frame and generate a video summary. The application further explores the influence of the spatiotemporal features extracted by 3D feature extractors with different spatiotemporal complexities on the video summary result. The application shows excellent performance on the video summary datasets SumMe and TVSum. Whether from the application scene or the performance index, the application has strong practical value.

The invention discloses a self-service guide explanation system and method based on mobile phone NFC, and belongs to the technical field of intelligent services of museums and exhibition venues. The system is composed of an NFC tag array, a WeChat applet client, a cloud management server and a content distribution network. A user holds a mobile phone to touch an NFC label close to an exhibit, so that a WeChat applet can be automatically opened, 'e touch-to-play 'can be achieved, label identification, content pulling and multimedia playing can be completed in a zero-configuration mode, and offline and online mixed display of audios and videos, pictures and texts and AR / VR is supported; meanwhile, the system records user behaviors and generates a personalized visiting report. The method solves the problems that in the prior art, special equipment needs to be rented, operation is complex, interaction is single, and fusion with a social platform cannot be achieved, and has the advantages of being low in cost, fast in deployment, zero in learning cost, rich in experience, easy to maintain and the like.

The application relates to a fine-grained traffic classification method based on an improved residual convolutional network in an SDN environment and belongs to the software technical field. In order to provide finer-grained application-aware traffic classification, let network operators better analyze network composition and manage and schedule network resources, fine-grained classification of the specific application programs is very important. Traditional methods tend to classify traffic based on protocols, which is coarse-grained classification. Inspired by the research in computer vision, the method of the residual convolutional network is applied to the identification and classification of network traffic. The method solves the network degradation problem that occurs in the process of fine-grained network traffic identification by traditional deep learning methods with the increase of network depth, can effectively learn deeper network features, and further realizes fine-grained network traffic classification.

The invention provides a matching generation method, device and equipment of a passive device network and a network group, a medium and a computer product, the passive device network is generated according to a specified matching parameter value, and the passive device network is composed of a first passive device network and a second passive device network which are connected in series or in parallel; the first passive device network and the second passive device network are both composed of a plurality of passive devices of the same type; based on the distribution condition of the equivalent electrical parameter values, determining a corresponding high-density distribution range; the matching parameter values are decomposed through the high-density distribution range; the parameter parts falling into the range are divided into decimal parameters, and the rest are divided into integer parameters; the first passive device network is provided with unit passive devices with an integer parameter number; and searching a series-parallel network matched with the decimal parameter, and constructing a second passive device network. The problem that in the prior art, an automatic tool is lacked to efficiently and accurately design a passive device network with overall equivalent parameters meeting expected design requirements is solved.

The application discloses a network space simulation construction and analysis display method, which comprises the following steps: editing and setting multiple types of network entity models and network relationship models, storing the network entity models and the network relationship models into a database, outputting the network entity models and the network relationship models from the database, constructing corresponding network simulation entities and a simulation target network, and performing hierarchical display on the simulation target network and the network simulation entities. The method can effectively realize simulation construction of the network space, can display the network architecture as a whole and display network composition details, and is favorable for intuitive network analysis and feature presentation.

The invention discloses a nonlinear system response prediction method based on a time sequence convolutional network and a main and auxiliary double-network structure, and belongs to the technical field of nonlinear system modeling and prediction. The method comprises the following steps: constructing a training data set, processing an excitation sequence and a response sequence of a nonlinear system through a sliding window method, and generating a plurality of samples; a main prediction network is constructed, the main prediction network is composed of a main time sequence convolution sub-network and a main condition sub-network, and the main condition sub-network outputs condition parameters used for carrying out condition normalization modulation on the time sequence convolution sub-network so as to directly output a future response prediction sequence; training the main prediction network to minimize the prediction error; the topological structure of the auxiliary correction network is the same as that of the main network, but parameters of the auxiliary correction network are independent, and a sequence obtained after main network prediction residual normalization serves as a training target for training; in the prediction stage, new excitation and historical response are input into the trained main network and auxiliary network, main prediction output and auxiliary output are obtained, and a final prediction result is obtained through formula calculation. According to the method, through time sequence convolution, condition normalization and cooperative work of the main network and the auxiliary network, the precision and robustness of nonlinear system response prediction are effectively improved.

The application discloses a kind of based on wireless self-organizing network topology's blockchain consensus method.According to the current topology of wireless self-organizing network, relay broadcast link is constructed;According to relay broadcast link, reliable broadcast network and asynchronous binary consensus network are constructed, and blockchain network is composed;According to asynchronous binary consensus network, consensus group is constructed;Periodically, the content proposed by each node is consensus, including two types of user data and network topology change information;After consensus, report to application layer, or update local network topology and re-perform system construction.The method inherits the decentralization characteristics of blockchain in network layer, avoids single point failure problem;Using the broadcast nature of wireless communication, an efficient broadcast mechanism based on network topology is used, which reduces the communication overhead in the blockchain consensus process;A blockchain construction mechanism based on network topology is used, which solves the mismatch problem of low dynamicity of traditional private chain consensus algorithm and high dynamicity of wireless self-organizing network.

The application discloses a kind of multi-graph hierarchical road network representation methods based on graph neural network, in the method, the structure perception graph neural network of spectral clustering and graph attention network is formed, modeling is carried out to road network representation of different levels, two kinds of virtual nodes, namely structure area and functional area, are introduced, multi-graph mechanism is used to guide virtual node to correspond with the structure area and functional area of real world, structural similarity is established using road type attribute, functional similarity between road segments is defined using city POI information, message sharing is then performed at high level, then updated information is propagated to low level nodes, and functional attributes of road network are supplemented;Using the method, road network representation containing structural features and functional features can be obtained;It is convenient to determine road network structure and functional role;It is helpful to reveal the functional area of city, helpful to route planning and time of arrival estimation and position prediction, and conducive to the construction of intelligent transportation system.

The invention belongs to the technical field of image processing, and relates to a self-priori-guided image restoration method and system, and the method comprises the following steps: S1, constructing a damaged image data set; step S2, a step of network construction; step S3, a step of network training and optimization; step S4, a network performance test step; according to the technical scheme of the invention, aiming at the problems of content repetition, color distortion and the like which are easy to occur when a traditional image restoration method is used for processing large-area defect of a continuous region or reference content is insufficient, a two-stage image restoration method consisting of a semantic prior estimation network and a semantic prior guide restoration network is adopted; and residual semantic information in the damaged image is fully mined and utilized, so that stronger robustness is realized while the visual naturalness and detail accuracy of a restoration result are improved.

This invention relates to a robust image tampering localization method based on consistency-guided learning, belonging to the fields of image content tampering forensics and deep learning technology. The method employs an alternating training strategy based on epochs, constructing a consistency-guided learning framework consisting of a target network and a source network: the target network takes the unprocessed image as input and performs joint optimization through localization loss and contrast loss; the source network takes the post-processed image as input and learns by fitting the response of the target network. This invention improves the model's robustness to post-processing operations while minimizing interference with the model's learning of tampering traces.

The present disclosure relates to the field of computer vision image dehazing and discloses a single image dehazing method based on detail recovery. The method includes step 1: constructing a training dataset; step 2: constructing a backbone dehazing network to achieve preliminary image dehazing: based on the U-Net network model, introduce a residual module based on pixel attention mechanism in the encoding region, and an feature enhancement module in the decoding area; step 3: constructing a detail recovery network for image detail recovery, and introducing residual shrinkage module and spatial attention mechanism; step 4: training an overall network model composed of the backbone dehazing network and the detail recovery network; step 5: Testing. The present disclosure can effectively remove fog while improving the problem of edge detail information loss, reducing the blurring of the edges of the dehazing image, and generating a higher-quality dehazing image.

The application belongs to the technical field of decoding, and discloses an LDPC decoding system based on dynamic hypernetwork and two-stage learning, which is composed of a main decoding network and a dynamic hypernetwork, the dynamic hypernetwork is composed of a plurality of hypernetwork units corresponding to each layer of the main decoding network one by one, and the dynamic hypernetwork and the main decoding network are a controller-executor relationship; the dynamic hypernetwork dynamically outputs key parameters of the main decoding network for each iteration, and the main decoding network uses the key parameters to complete message passing and LLR updating; the input of the first hypernetwork unit is a log-likelihood ratio (LLR) value vector received by a channel, and the vector carries key information of a received signal; starting from the second hypernetwork unit, the input of the tth hypernetwork unit is the output of the t-1th layer of the main decoding network. The application significantly improves decoding performance, greatly improves decoding efficiency, enhances environmental adaptability, and reduces calculation and hardware costs.

A system that comprises an IP-routed interregional distribution network, each comprised of two or more IP-routed regional distribution networks (RDN), each of which employs two or more user-network interfaces (UNIs), each including (a) a physical network including any device capable of operating at network layers 2 and 3, (b) a first virtual broadcast domain (VBD), (c) a second VBD, (d) a regional virtual extensible local area network (VXLAN), and (e) a protocol transformation stack (PTS). The UNI is adapted for layer 2 connection to a user device via the first VBD, and adapted for layer 3 communication over the IP-routed RDN via the VXLAN. The PTS is adapted to convert a layer 2 broadcast domain to / from an IP-routable form by mapping the first VBD to the second VBD, encapsulating the second VBD into the regional VXLAN, and translating that VXLAN into a second interregional VXLAN.

This invention belongs to the field of remote sensing image processing and deep learning technology, specifically involving a refined training method for YOLO deep learning models. The method includes acquiring a remote sensing image dataset, preprocessing it and initializing the basic network, calculating the detection probability of each target category in the preprocessed remote sensing image, selecting target categories suitable for the current network based on the detection probabilities, and decoupling these target categories. Based on the weights of the current network, the network is dynamically expanded so that the expanded network's feature extraction layers can provide suitable feature learning capabilities for the undecoupled target categories. This process of network expansion and target category decoupling is iteratively performed during training until all target categories match the optimal network, ultimately forming an ensemble model composed of M networks with different number of layers. H =[ H 1 K1 , H 2 K2 ,..., H M Km ],in K m Let m be the feature extraction layer number of the m-th sub-network. This invention allows for flexible adjustment of parameters such as the initial network layer number and the number of extended layers, making it suitable for various target detection tasks.

The application discloses an anti-knowledge forgetting mutual guidance type semi-supervised medical image segmentation method, first, a double network model composed of two initialized different networks is constructed, and the internal difference is used as an implicit disturbance source; secondly, a double-path data fusion strategy is designed, structured fusion of region blocks of labeled and unlabeled images is carried out, a forced model is generated, and mixed images of two data domains are simultaneously processed in a single iteration, so that forgetting of learned labeled knowledge in the semi-supervised learning process is effectively inhibited; then, a mutual guidance learning mechanism is adopted, the prediction result of one network is used to generate high-quality pseudo labels, and a unified mixed supervision signal is constructed for the mixed images; finally, the model is trained by jointly optimizing a loss function customized for the mixed samples. The application is suitable for two-dimensional and three-dimensional image segmentation tasks, can significantly improve segmentation performance and robustness in a small amount of labeled data scene, and can enhance the universality and ease of use of the method.

The invention discloses a digital economic index prediction system and method based on big data, and the system comprises a multi-source data collection module which is used for obtaining digital economic related original data from a plurality of data sources; the data preprocessing module is used for cleaning, converting and standardizing the original data to obtain preprocessed data; the digital economic feature engineering module is used for extracting feature indexes from the preprocessed data; the MoE prediction model module is composed of a gating network and a plurality of expert networks and is used for performing digital economic index prediction based on the characteristic indexes; the Bayesian optimization module is used for dynamically adjusting hyper-parameters of the MoE prediction model; and the model evaluation and feedback module is used for evaluating prediction performance and optimizing a feature selection strategy. The invention belongs to the technical field of big data processing, and particularly provides a method for solving the problems of single data dimension, insufficient model adaptability, lack of multi-scale analysis and lack of real-time optimization capability in the prior art.

This application pertains to the field of spacecraft on-orbit servicing. It provides a robust model prediction and efficient configuration characterization method for space rigid-flexible coupled systems. The model prediction neural network in this disclosure consists of an LSTM encoder and a fully connected network with three hidden layers; the final output of the neural network is the net depth. By capturing the state variables of MUs in a past time series and the temporal mapping relationship between the net depth and the net depth at the next time step, an affine dynamic equation implicit in the net depth is established. Based on this, a cost function coupling formation and configuration information is constructed. The required control gain matrix is ​​obtained using a system identification method. Adaptive dynamic programming is used to approximate the optimal solution of the Hamiltonian function corresponding to the cost function, iteratively updating the network weight update rate during the process. Finally, the optimal control input is applied to the corresponding MU to achieve coordinated control of the TSNR system configuration and formation.

A new generation of network resource management architecture for coverage is provided, which constructs a structure of “signal coverage+capacity coverage” by incorporating satellite, space-based, and ground access points to achieve large-scale regional signal coverage and local area capacity coverage. In the core network, a global resource orchestrator is added, and in the access network, local resource orchestrator and virtual network management and maintenance unit are added for multiple functional units of resource management and control. In terms of coverage structure, it is divided into three-dimensional dense capacity coverage and three-dimensional global signal coverage. The former consists of ground dense network and aerial ad-hoc network, respectively realizing high traffic density coverage and large connection three-dimensional coverage. The latter consists of ground network and satellite constellation, respectively realizing extended coverage towards low altitude and three-dimensional coverage towards space, sky, ground, and ocean.

The invention discloses a profit maximization Canji air transportation network design method under government subsidy, which is used for improving the benefit of an airline company and perfecting an air transportation system. The method comprises the following steps: firstly, analyzing composition and structure of a trunk-branch air transportation network; and then realizing accurate prediction of the aviation travel demand based on a composite measurement economics model. Secondly, a main branch air transportation network optimization model considering government subsidy and profit maximization is established, and the model integrates mixed distribution, direct connection, capacity limitation and dynamic demand response mechanisms and is used for analyzing the coupling influence of a pricing strategy, a subsidy scheme and the like on the profit. And finally, proposing a three-layer nested heuristic algorithm, and generating an optimal network design scheme by progressively solving network structure optimization, route subsidy scheme matching and path distribution decision. According to the method, the airline layout of an airline company can be optimized, the government subsidy utilization efficiency is improved, and the sustainable development of general aviation is promoted.

The invention discloses a metasurface robust reverse design method for a complex interference scene, and relates to the crossing field of metasurface reverse design and a deep learning technology. The method comprises the following steps: 1) constructing a data set containing metasurface phase distribution and an ideal far-field pattern; 2) constructing a series neural network architecture consisting of a global cross attention Transform reverse network and a pre-trained forward prediction network; 3) training the GCAT by using the training set only containing the non-interference data; and 4) inputting interfered actual far-field data into the trained GCAT network, and outputting target metasurface phase distribution. According to the method provided by the invention, high reconstruction fidelity can be kept in various complex interference scenes without training on interference data, the method has the advantages of high robustness, excellent generalization ability, high design efficiency and the like, and the problems of low reverse design precision and poor generalization ability in a complex electromagnetic environment in the prior art are effectively solved.

The application discloses a UAV target tracking method based on feature fusion and online template updating, and comprises the following steps: a deep network model is constructed based on a Resnet residual network and a hollow convolution, target features can be effectively extracted, the effective receptive field of the features is enhanced, the detailed features of a shallow network and the semantic features of a deep network are efficiently fused, and the expression capability of the features is enhanced; a template branch and a detection branch in a Siamese structure are composed of a ResNet-50 network and a target fusion network, feature maps of the two branches are sent into a cascaded cross-correlation module, and the target position is determined; a template library is constructed, a response map score is calculated, and the template branch is updated online according to a threshold value. The application reaches a high level in terms of tracking success rate and accuracy, and effectively improves the UAV target tracking performance.

The invention discloses an AGV cluster dynamic path planning control method and system based on reinforcement learning, and the method employs a centralized training distributed execution architecture, and comprises the steps: defining an observation state space and a discrete motion space which comprise the position, the target position and a local environment map for each AGV; designing a collaborative reward function fused with global conflict prediction; a multi-agent reinforcement learning model composed of an actor network and a reviewer network is constructed, the actor network outputs action probability distribution according to local observation to control the AGV to act, and the reviewer network evaluates a joint action value by using global information to guide training; training the model in an off-line manner in a simulation environment until convergence; the trained actor network is deployed to each AGV, and online distributed real-time path planning is realized; aGV cluster deadlock and collision can be effectively avoided, and system operation efficiency, robustness and dynamic environment adaptability are improved.

This disclosure provides a training method and apparatus for an object detection model, an object detection method and apparatus, a computer-readable storage medium, and an electronic device. The training method for the object detection model includes: using a backbone network composed of at least two sub-networks within the object detection model to be trained, performing step-by-step feature extraction on a sample image to obtain feature data, wherein the at least two sub-networks are arranged in order of data processing volume; using the detection network within the object detection model to be trained, obtaining the detection result of the target object in the sample image; determining a loss value representing the error between the detection result and pre-annotated information for the target object; and adjusting the parameters of the object detection model to be trained based on the loss value. This disclosure helps to more accurately detect target objects from images because the lower-level sub-networks can acquire finer-grained features from the image.

The application designs an optimization method suitable for a deep learning network for a classification task. Existing deep network optimization methods mainly use a large network as a teacher network to migrate knowledge to a student network. However, the Soft target is the Softmax output of the teacher network, and the Hard target is the dataset label, which is not suitable for a distributed student network. The application retains the basic framework of Teacher-Student knowledge distillation, but changes the Soft target to the feature output of the last convolutional layer of the Teacher network, comprehensively evaluates the influence degree of feature points on the final result according to the intra-class and inter-class fluctuations, and then filters and samples the feature points according to the composition of the distributed network. Finally, the spatial fluctuations of the Teacher-Student are counted into the loss function, and the overall training of the network is performed. The application can distribute the deep learning model to an embedded terminal in a distributed manner, and lays a foundation for the landing of deep learning.