56 results about "Classful network" patented technology

The invention discloses an unknown attack detection method and system for intelligent network security situation awareness, and the method comprises the steps: obtaining multi-dimensional data of a power grid monitoring system, carrying out the data preprocessing, carrying out the network attack preliminary detection of the multi-dimensional data through a pre-built network security situation awareness framework, and outputting suspected attack data, the method comprises the steps of performing data training on historical attack data through an improved OCN open set classification network, identifying attack features of each known attack type, performing semantic similarity calculation on the attack features and suspected attack data, performing attack feature mapping and clustering on the suspected attack data based on semantic similarity, and obtaining an unknown attack feature clustering result. And according to the unknown attack clustering result, carrying out attack type classification on unknown attacks in the suspected attack data to obtain an unknown attack detection result. The method has the effects of detecting unknown attack means in time, effectively reducing the risk that the power grid system suffers from network attacks and guaranteeing safe and stable operation of the power grid system.

The invention discloses an inverter IGBT intelligent fault diagnosis method based on a multi-scale attention network, and relates to the technical field of power electronic equipment fault diagnosis. A three-phase current signal of a three-level NPC inverter is collected and preprocessed; multi-dimension fault features of the current signals are extracted in parallel through multi-scale depth separable convolution; a scale-aware efficient attention mechanism is introduced to carry out weighted optimization on each scale feature, and key fault information is strengthened; and outputting a fault diagnosis result through a feature fusion and classification network, and performing INT8 quantization processing on the model to adapt to an embedded deployment scene. The method has the advantages of being high in multi-scale feature capturing capacity, high in diagnosis precision and light in model weight, and can meet the dual requirements of industrial application for real-time performance and accuracy.

The invention relates to a federal semi-supervised domain adaptive time sequence learning method, which comprises the following steps: aiming at a target classification network formed by a coding feature extraction layer and a classifier head, firstly, executing pre-training by a server based on local label data, freezing the classifier head, and then, respectively executing pre-training by each client based on local label-free data; unsupervised training is carried out on the coding feature extraction layer, federated learning is realized and a mobile behavior recognition model is obtained in combination with fusion of each trained parameter of the local coding feature extraction layer of each client by the server, and each preset mobile perception data acquisition is analyzed and a corresponding mobile behavior label is output, so that mobile behavior perception application is realized. The design method not only can improve the classification accuracy of the target classification network on the client data, but also lays a solid foundation for promoting the wide application of semi-supervised federal learning in a complex real scene.

The present invention is an unsupervised integrated method of instability detection and fault module positioning, including: acquiring a topological structure of a direct current (DC) microgrid and collecting electrical data of each node in the topological structure to construct a corresponding first enhancement dataset and a corresponding second enhancement dataset; constructing a fault type pool based on the topological structure; constructing a corresponding classification network based on a twinborn network framework; training the classification network using the prepared datasets to obtain a detection model; and inputting the electrical data of the DC microgrid to be detected to a detection model, to output whether the DC microgrid to be detected has system stability and a corresponding fault type. Further provided in the present invention is an unsupervised integrated device of instability detection and fault module positioning.

The invention discloses an AI-driven vehicle-mounted sound field real-time modeling and voice separation method, and relates to the technical field of voice signal processing. A main control unit comprising a time sequence synchronizer, a resource scheduler and a health monitor is constructed. 3D sound field modeling is carried out by adopting a lightweight STCN + bidirectional LSTM network, adaptive updating of the model is realized through EWC incremental learning, and a CNN-LSTM noise classification network and targeted suppression algorithms such as ANF / spectral subtraction are developed. The voice separation module adopts an improved Conv-TasNet architecture, 3D spatial constraint and a multi-task loss function are fused, and low delay is realized under INT8 quantization and pipeline processing. The system dynamically optimizes parameters through a real-time regulation and control unit, supports scene self-adaption, finally achieves a separation effect in a mixed noise scene, reduces the delay of the whole system, and effectively improves the definition and stability of vehicle-mounted voice interaction.

The invention relates to the field of microservice system anomaly detection, in particular to a microservice anomaly detection method and device. The method comprises the steps of obtaining historical data of a micro-service system, and obtaining first modal data and a log time sequence matrix based on historical service index data, historical log data and historical call chain data in the historical data; obtaining a first modal feature matrix and a second modal feature matrix according to the first modal data and the log time sequence matrix; obtaining a historical anomaly detection result according to the first modal feature matrix and the second modal feature matrix; obtaining first and second modal reconstruction data according to the first and second modal feature matrixes; optimizing the language model and the classification network to obtain an optimized language model and an optimized classification network; and preprocessing the collected real-time data, and inputting the preprocessed real-time data into the optimized language model and the classification network for anomaly detection to obtain a real-time anomaly detection result. The method is used for realizing accurate positioning and detection of the abnormity of the micro-service system.

The embodiment of the specification provides a multi-party participated model training method, system and device, the method comprises: a first party inputs a plurality of first features into a first feature extraction network, homomorphically encrypts an output result to obtain a plurality of first encrypted vectors and sends the plurality of first encrypted vectors to a second party; the second party inputs a plurality of second features into a second feature extraction network, homomorphically encrypts an output result to obtain a plurality of second encrypted vectors; the second party homomorphically calculates the plurality of first encrypted vectors and the plurality of second encrypted vectors to obtain a plurality of fusion encrypted vectors and sends the plurality of fusion encrypted vectors to the first party after being disordered, and records a first correspondence before and after disordering; the first party decrypts the plurality of fusion encrypted vectors and inputs the plurality of fusion encrypted vectors into a first part of a classification network to obtain a plurality of latent vectors; the first party uses the plurality of latent vectors and a classification label, and the second party uses network parameters of a second part of the classification network and the first correspondence to perform multi-party secure calculation and determine a first loss; the first party and the second party update the classification network according to the first loss.

The invention relates to the technical field of federated learning algorithms, in particular to a fair federated learning method based on multi-agent reinforcement learning, which comprises the following steps: respectively initializing network parameters of a server and a local client, and calculating an initial state vector of a local environment by the local client according to the initialized network parameters; the server calculates the initial state of the global environment, and a local client selects and executes an action; the server updates the classification network and broadcasts the classification network to all local clients, and the local clients update local states and rewards; the server updates the global environment state and stores the data in an experience batch; and the server updates the value network, and the local client updates the strategy network. The strategy network is trained based on centralized training and distributed execution, and the convergence speed of the strategy network can be greatly improved.

The application discloses a multi-modal radio frequency authentication method based on a multi-scale signal representation, which comprises the following steps: firstly, pre-processing the original IQ signal of a target device received to obtain an instantaneous envelope signal; carrying out multi-scale decomposition and denoising processing on the instantaneous envelope signal to obtain a denoised envelope signal; constructing a multi-modal data set according to the original IQ signal and the denoised envelope signal; carrying out feature extraction and fusion on the multi-modal data set through a pre-trained target feature extraction network to obtain a fused feature representation; and carrying out classification on the fused feature representation through a pre-trained target classification network to output a classification result corresponding to the target device. The amplitude, frequency and time-frequency energy distribution information can be complementarily fused by constructing the multi-modal data set; the multi-modal features are extracted and fused through the target feature extraction network, so that the information loss is effectively avoided; and the target classification network is used for rapid classification, thereby reducing the calculation cost and ensuring the accuracy of identification.

The application discloses a kind of multi-modal perception data computer system fault diagnosis method and system, the method of the application includes the unified vectorization characterization of the multi-modal perception data of computer system, for the data of each modality in multi-modal perception data, respectively utilize the expert network of multiple channels to generate vectorization feature, according to the data modality of input, share routing network and the modality routing network of corresponding modality are cascaded to calculate the trust degree of the data of one modality to different expert networks, and the feature representation of this data modality is obtained based on trust degree to each expert network feature output weighted summation;The feature representation of each modality is characterized to realize superposition;The feature obtained after feature confusion is sent into classification network to obtain the fault diagnosis result of computer system.The application aims to comprehensively use multi-modal perception data to realize the fault diagnosis of unified computer system, to solve the inconsistency problem when different modal independent fault diagnosis.

The present application belongs to the field of computer vision, and particularly relates to a weakly supervised object localization method based on category correction. In order to solve the problem of inaccurate localization of the CAM technology, the rough-to-fine process is adopted instead of using the category feature map for localization. The model of the present application is composed of a backbone network, a localization network and a classification network. First, the localization network generates a class-independent segmentation map by using unsupervised segmentation technology, so as to determine the rough position of the target object. Then, the classification network is used for fine-grained correction through the category label. The method based on category correction can accurately locate the object and well identify the contour details.

The application discloses a DAS signal cross-scene multi-category classification method, system, device and medium, belongs to signal recognition classification in the field of optical fiber sensing technology, and aims to solve the technical problems that the DAS signal can only recognize single scene events in the prior art, and the DAS signal cannot be recognized and classified in a complex scene. The application comprises the following steps: acquiring samples and labels; constructing a signal recognition classification model comprising a feature extraction network and an identification classification network; the identification classification network comprises a tree classifier, each non-leaf node of the tree classifier comprises a node classification sub-network and an output layer, the node classification sub-network comprises a one-dimensional convolution layer, a batch normalization layer, a ReLU layer, a one-dimensional maximum pooling layer, a one-dimensional convolution layer, a batch normalization layer, a ReLU layer and a one-dimensional maximum pooling layer, and the output layer comprises a transformation layer, a full connection layer, a ReLU layer, a full connection layer, a ReLU layer and a Softmax layer arranged in sequence; training the signal recognition classification model; and classifying signals in real time.

The present application relates to a kind of based on reinforcement learning's unsupervised mismatch detection method, the present application first inputs the matching point without label to mismatch detection network, training detection network, by multiple iterations to obtain best network parameter, and using the final network model of saving best network parameter to the matching point set to be detected for detection, by generating N extract subset and evaluating its corresponding model, finally the maximum consistent set of model is regarded as correct matching point pair set, to complete mismatch detection and elimination.The present application has higher mismatch detection efficiency, precision and stability, and on the one hand, it overcomes the sample label problem;On the other hand, it is not limited by mismatch rate, and a large number of correct matching points can be obtained with fewer sampling times;And the method is a kind of unsupervised learning framework, which can be compatible with other classification networks, to solve the mismatch detection problem.

The invention discloses a power consumption data anomaly detection method and device based on decoupling anomaly injection driving. The method comprises the following steps: injecting predefined multiple types of pseudo anomalies into a normal data window to generate a pseudo anomaly window; the method comprises the following steps: pre-training a pre-constructed submerged space hybrid anomaly expert network model architecture, and generating a shared encoder, a normal decoder, an anomaly decoder and a plurality of anomaly expert networks of a hybrid anomaly expert network of a submerged space hybrid anomaly expert network model; learning a routing network and a classification head of the hybrid anomaly expert network, and generating a routing network and a classification network of a submerged space hybrid anomaly expert network model; calculating a reconstruction error of to-be-tested data at each time point and an anomaly probability output by the classification network based on a sliding window by adopting a submerged space hybrid anomaly expert network model, and performing linear combination to obtain a point-by-point anomaly score; and judging the point-by-point anomaly score according to a set threshold value to obtain an anomaly detection result.

The application discloses a high-speed moving target recognition method based on data enhancement and belongs to the field of target recognition. The application introduces a high-speed moving target background data set, the confidence of a pseudo target obtained by a classification network after the pseudo target generated by a differentiable generative adversarial network, and the target instances with a confidence greater than a set threshold and the target instances in the initial data set are segmented objects and are enhanced together in the high-speed moving target background data set, solving the problems of insufficient samples and inconsistency between the high-speed moving target training background and the working environment. When the pseudo target generated by the differentiable generative adversarial network is trained, the confidence of the pseudo target and the CIoU value of the positive sample boundary box and the real box are weighted and summed to form a new confidence loss function in the YOLOv7 target detection algorithm, and the improved loss function can more accurately measure the authenticity of the target. The method can realize accurate recognition of a specific type of high-speed moving target under small samples.

The application discloses a kind of offshore ship target detection methods based on generation confrontation network anti-shake, comprising the following steps: establishing DeblurGAN-v2 network model and training, using generator and discriminator alternately training method training, finally loss function converges and obtains the deblurred confrontation generation network model;Take out the discriminator network part in the trained network model as binary classification network continues training;Establish based on offshore ship target detection model and training, obtain YOLOv7 detection model;The generator part in the obtained trained network model, binary classification network and YOLOv7 detection model are reformed and fused, new discriminator network is used as front-end module, for fuzzy frame judgment, for selecting whether the picture sample is sent into generator network to deblur, and the target detection network of fusion generation confrontation network DeblurGAN-v2 is constructed.

The embodiments of present disclosure address unresolved problems of label inconsistency, where outputs of different levels create impossible combinations, and error propagation from previous level outputs can significantly impact its performance. Embodiments provide a method and system for a Progressive Multi-level Training framework with a Logit-masking strategy (PMTL) for a retail taxonomy classification. PMTL enables neural network models to be trained separately for each level to reduce error propagation problems. To further enhance the model's performance at each level and get the label-wise constraint from the previous level, the global representation from model of previous level is augmented. Further, a logit masking strategy is used to restrict model(s) to learning only relevant classes through part of final classification layer, thereby addressing label inconsistency issue, and incorporating benefit of parent node-based local classifier. This framework is generalized irrespective of dataset size and is configured for attaching to any hierarchical classification network.

The invention discloses a Spine-Leaf network hidden fault detection method based on deep learning, and the method comprises the steps: collecting the structure information of a Spine-Leaf network and the time sequence state data (TX, RTT and Qlen) of an RDMA network card, analyzing the feature rule of the time sequence state data of the RDMA network card when hidden faults exist, building a feature encoder based on LSTM and CNN dual-branch fusion and a classification network based on a softmax function, and carrying out the detection of the hidden faults of the Spine-Leaf network. And then training model parameters by using a cross entropy loss function in an offline training stage, and finally outputting a probability that a communication path has a hidden fault based on real-time state data of the terminal RDMA network card in an online evaluation stage. The method can effectively solve the problems that Spine-Leaf network hidden fault feature capture is not comprehensive and detection is not sensitive, hidden fault recognition accuracy and real-time performance are improved to the maximum extent, and efficiency loss of a distributed deep learning task caused by network abnormity is reduced.

The invention relates to the technical field, in particular to a DNS (Domain Name Server) tunnel attack real-time detection method based on LSTM-Transform hybrid architecture, which comprises the following steps of: 1, acquiring continuous DNS data streams in a network, and extracting an original DNS query sequence as an input feature; 2, constructing a hybrid neural network architecture composed of a time sequence feature extraction module and a global dependency modeling module; 3, an information fusion module fuses the time sequence feature vector and the long-distance dependency feature vector to generate a joint representation vector; and step 4, the classification network model performs attack behavior classification based on the joint representation vector, and outputs a detection result of DNS tunnel attacks in real time. According to the method, the deep time sequence rule and the global dependency feature of the DNS query sequence are cooperatively extracted through the LSTM-Transform hybrid architecture, and the inherent DNS protocol behavior pattern of the network node can be accurately modeled.

This invention relates to the field of deep learning technology, and more particularly to a dynamic clustering channel pruning method based on reinforcement learning. The method involves pre-training a classic classification network using training images to obtain a pre-trained network model; statistically analyzing the multi-frequency responses of the pre-trained network model to construct a channel-level network state representation; constructing a hypergraph using the network state representation as vertices, and aggregating information from the vertices in the hypergraph using hypergraph convolution to obtain an updated hypergraph; planning the pruning method for the pre-trained network model based on the updated hypergraph; employing reinforcement learning to prune the pre-trained network model according to the pruning method, and then retraining the pruned network using the training images to obtain the final network model. This invention effectively identifies redundant channels and improves pruning performance by combining statistical analysis of channel responses across multiple frequencies for network state representation, the ability of hypergraph convolution to integrate complex information, and reinforcement learning.

The application discloses a method for obtaining a network model, a storage medium and an electronic device. The method comprises the following steps: obtaining a target data set; constructing a second network model based on a first network model and a dynamic routing structure, wherein the first network model is an image classification network model, the first network model comprises a plurality of network substructures, the dynamic routing structure is used for selecting to skip or execute at least one network substructure in the plurality of network substructures, and the second network model is an initial dynamic inference network model based on path selection; training the second network model by using the target data set to obtain a third network model; and performing binary processing on the dynamic routing structure based on the third network model to obtain a fourth network model. The application solves the technical problem of low learning efficiency, low flexibility and large calculation consumption of the network model in the related art.

The application discloses a contract attack detection method and device based on symbolic execution and a graph neural network, comprising: obtaining deployment bytecode and runtime bytecode of a smart contract; performing symbolic execution on the runtime bytecode, extracting external calls and event information through depth-first search and taint tracking; constructing a contract call information graph and iteratively simplifying it; performing multi-dimensional feature coding on each node to generate a node feature vector; encoding on-chain metadata of a deployer into a deployer feature vector; constructing and training a graph embedding network and a classification network, the input of the graph embedding network being the simplified contract call information graph, the node feature vector and edge type information, and the output being a graph-level feature vector; the input of the classification network being a joint feature vector obtained by splicing the graph-level feature vector and the deployer feature vector, and the output being a contract classification result. The application can cover multiple attack types and realize attack contract identification with low false alarm rate by extracting contract execution logic structures for detection.

The application discloses a building method of an image classification network with a fusable multi-branch architecture, and belongs to the technical field of computer vision. A series of convolution kernels of different sizes are used to capture multi-scale information in an image to obtain better performance. Then, three fusion algorithms (fusion algorithms 1, 2 and 3) are proposed to fuse the convolution kernels, the normalization layers in the convolution kernels, the activation functions, the residual edges and the convolution kernels in multiple aspects, so that the multi-branch architecture is fused into a single-path architecture, and the reasoning speed of the network is accelerated without reducing the accuracy.

A three-stage gold wire bonding defect detection method, characterized by comprising three stages, as follows: (1) chip and key connection area segmentation: using DeepLabV3+ model to complete semantic segmentation of the chip and key connection area; (2) interested region extraction: according to the different bonding wire connection objects and the complexity of the background, the gold wire bonding picture is divided into four categories, based on the position information extracted in stage 1, further design corresponding image processing algorithm to complete the interested region extraction; (3) defect discrimination: based on the extracted interested region, using image processing algorithm and a new classification network based on twin network architecture is proposed to complete the defect discrimination. Therefore, the present application is a three-stage defect detection method which combines deep learning method with traditional digital image processing method, and completes the defect detection of missing wire and bent wire in the gold wire bonding process of microwave assembly.

The application discloses a kind of based on IDC-Net network's ceramic tile product surface defect classification method and system, belong to ceramic tile surface defect classification technical field, including.The application designs the basic unit IDC-Block with feature enhancement capability, the module integrates dynamic large kernel convolution DLK and inherits the advantage of efficient feature extraction of visual state space block VSSBlock IDC-Layer is constituted, on the basis of IDC-Layer, dense connection design is used to constitute IDC-Block by multilevel structure stacking, gradually enhanced feature representation ability;Further design adaptive gaussian pooling module AGP, reduce the information loss in down-sampling process, significantly improve the classification performance of model to ceramic tile small target defect;Dynamic feature fusion module DYFF is designed, adaptive fusion of different resolution and semantic level features is realized, and the robustness of feature expression is enhanced;With EfficientNetV2-s classification network framework as auxiliary branch, local and global information is integrated to constitute double-path classification network IDC-Net by DYFF module, and the classification accuracy is further improved.

The application discloses a method and system for incremental classification learning of target objects in a dynamic service scene, and new class data sets and old class memory network sets at T time are input into a classification network, the new class data sets are used for training the classification network to realize incremental learning, and the old class memory network sets are used for a replay process of the classification network at T time to prevent catastrophic forgetting; through training, parameters of the classification network are updated to obtain a classification network capable of identifying new class and old class objects; during training, a prototype selection strategy based on density peaks is used to select a prototype for each new class in the new class data sets at T time to obtain a prototype set corresponding to each new class; based on the new class data sets and the prototype set corresponding to each new class, a self-organizing incremental learning neural network is constructed for each new class to obtain a new class memory network set; and the new class memory network set and the old class memory network set are merged to obtain an updated memory network set at T time.

This invention discloses a three-dimensional breast ABUS image classification method, electronic device, and computer-readable storage medium based on a tokenized bi-branch selective state-space model. The method preprocesses the three-dimensional ABUS volume data and inputs it into a hierarchical pyramidal classification network. The network constructs local convolutional branches and a global state-space branch in its basic modules: the local branch uses lightweight three-dimensional grouped convolution to extract texture and boundary morphology; the global branch aggregates features into a token map through voxel token generation, then performs multi-axis bi-directional selective state-space scanning and adaptive fusion using routing weights. After detoxing, the core features are injected through a gating mechanism. Finally, the classification result is output through three-dimensional global average pooling and a fully connected layer, enhancing the ability to model long-range dependencies across slices with near-linear complexity.

The invention relates to the technical field of network intrusion detection, and provides an unknown attack detection method and system in a label noise scene. The method comprises the following steps: step 1, acquiring a network flow data set with a label, and preprocessing the network flow data set to obtain a one-dimensional feature vector set with a label; 2, training a neural network by using the one-dimensional feature vector set, and in the training process, introducing a noise label correction mechanism to perform label correction on the one-dimensional feature vector set so as to train the neural network; wherein the neural network comprises a feature extraction network and a classification network; and step 3, inputting the network traffic to be detected into the trained neural network to obtain a detection result.

The application provides a power equipment state diagnosis method and device based on multi-source data fusion and a terminal equipment, and belongs to the technical field of power equipment state diagnosis. The method comprises the following steps: acquiring image, text and running sound data of power equipment, and inputting the data into a detection model comprising an image, text, sound backbone network, multi-modal feature fusion network and target classification network. Visual, semantic and audio features are respectively extracted by using each backbone network, multi-modal feature standardization vectors are generated, adaptive weight fusion is performed on the multi-modal feature standardization vectors by using the multi-modal feature fusion network to obtain multi-modal fusion features, and finally, the fusion features are identified by using the target classification network to output the type and state diagnosis result of the power equipment. The application can solve the problem of the existing technology that the detection accuracy decreases and misjudgment and missed judgment easily occur in a complex and low-light scene.