79 results about "Feature dependency" patented technology

The invention provides a multi-modal image fusion method based on modal self-adaption and modal interaction compensation, and the method comprises the following steps: S1, obtaining a multi-modal image fusion data set, and obtaining a training data set through preprocessing; S2, analyzing the modal difference characteristics of infrared and visible light images, and evaluating the correlation characteristics of image pairs in different scenes; s3, capturing a cross-modal feature dependency relationship through a self-attention mechanism; s4, a differential feature extraction strategy is adopted, model parameters are optimized through iterative training, and multi-modal image fusion is completed; s5, a modal interaction compensation module is additionally arranged, unit dynamic balance common features and modal exclusive features are fused, feature complementation is achieved in channel and space dimensions, parameters of the modal interaction compensation module are optimized, the model is made to learn the optimal fusion weight of the multi-modal features in a self-adaptive mode, and multi-modal fusion image generation optimization is achieved through the model; according to the invention, multi-modal image fusion can be accurately and effectively carried out.

The invention discloses an anomaly detection method based on a multi-mode denoising diffusion model, and belongs to the technical field of anomaly detection. Aiming at the problem that an existing anomaly detection method is easily influenced by single-mode information limitation and insufficient feature dependency relationship modeling in industrial quality inspection to cause a poor anomaly reconstruction effect, the method comprises the following steps: firstly, constructing a multi-mode large language model and a CLIP encoder, jointly generating an anomaly semantic description, and embedding a text mode prompt into a de-noising diffusion process; secondly, designing a denoising UNet network with a four-stage hierarchical structure, integrating a mixed attention enhancement Transform module in each stage, and synchronously capturing a long-range dependency relationship in space and channel dimensions and strengthening local details by performing traversal connection on channel-level self-attention and a pixel attention mechanism; and finally, realizing accurate reconstruction of the abnormal region through cross-modal feature alignment. Experiments show that the method improves the reconstruction quality of the abnormal image and the accuracy of anomaly detection, and has high robustness in a complex industrial scene.

The invention provides a network security malicious code binary search method and system, and relates to the technical field of data processing, and the method comprises the steps: calculating the multi-dimensional structural similarity between a to-be-detected binary file and each standard binary reference file based on a function level control flow semantic feature vector; determining a finally matched standard binary reference file according to the multi-dimensional structural similarity; performing fine-grained difference analysis on the binary file to be detected and the finally matched standard binary reference file to obtain a fine-grained difference analysis result; and on the basis of a fine-grained difference analysis result, in combination with sensitive data operation behavior feature detection, judging whether the to-be-detected binary file is a maliciously tampered version, and positioning a malicious code injection point. According to the method, the defects that in the prior art, dependence on fixed features is too high, and compilation optimization is sensitive are overcome.

The invention discloses a radiology report generation method and system based on global dependency learning, and belongs to the technical field of natural language processing. According to the method, a global dependency learning module is designed, and the module integrates rotation position coding to enhance original Mama, so that long-sequence visual feature dependency is effectively modeled, and the model calculation efficiency is improved. According to the method, the problems that the visual long-term dependence modeling and calculation efficiency in the radiology image are difficult to effectively balance, and the heterogeneity of the characteristics of different modal information of the image and the text is difficult to effectively align and fuse are solved; the method not only can effectively capture the long-term dependency relationship on the key visual features in the radiology image, but also can improve the model calculation efficiency, and can enhance the alignment and fusion capability among the multi-modal heterogeneous information.

The invention provides a special disease first-aid full-process auxiliary decision-making method based on multi-modal perception and AI, and relates to the technical field of medical first aid, and the method comprises the steps: obtaining first-aid field multi-modal data, constructing a feature dependence graph, and recognizing the type and symptom features of a special disease; indexing special disease diagnosis and treatment data based on the symptom feature vector, and determining a reference case and a treatment path; the action space is constructed in combination with clinical specifications, the expected income is predicted by applying Monte Carlo search, and the optimal treatment scheme is generated, so that the first-aid special disease recognition accuracy can be improved, the treatment decision time can be shortened, and the first-aid intervention effect can be improved.

The invention provides a high-precision phase unwrapping method fused with a Transform self-attention mechanism. The high-precision phase unwrapping method is suitable for digital speckle interferometry, optical imaging and medical image processing. According to the method, on the basis of a U-Net structure, position coding and multi-head self-attention (MHSA) modules are introduced into a bottleneck layer, so that global feature dependence is established, and a long-range context is captured; a multi-head cross attention (MHCA) module is added in a decoding stage, attention weighting and noise suppression are performed on jump connection features, and selective fusion of key features is realized. A composite loss function is adopted in training, error variance loss and total variation loss are included, and phase periodic characteristics and smooth constraints are considered. Through simulation data training containing noise and complex phase jump, the unwrapping precision and robustness are significantly improved.

The invention relates to the technical field of sales prediction, in particular to an electronic product sales data prediction method and system based on artificial intelligence, and the method comprises the steps: obtaining historical sales time sequence data of a to-be-predicted product; constructing an initial sales prediction model according to the historical sales time sequence data; based on one or more predefined market impact morphological functions, generating an adversarial training sample, the market impact morphological functions being used for simulating external events causing abnormal fluctuation of sales volume; performing adversarial training on the initial sales prediction model by using a training data set containing adversarial training samples to obtain a prediction model with enhanced robustness; and predicting the future sales volume of the to-be-predicted product based on the prediction model with enhanced robustness. Therefore, the problems of strong external feature dependence, weak generalization ability to unknown impact, single model structure and the like in the prior art are solved.

The invention discloses an image defogging method based on multi-scale feature fusion. The method comprises: obtaining a foggy image and dividing the foggy image into a training set and a test set; constructing an original multi-scale feature fusion image defogging model, and performing training evaluation by using the training set and the test set to obtain a multi-scale feature fusion image defogging model; and inputting a to-be-processed foggy image into the multi-scale feature fusion image defogging model, and outputting a defogged image. According to the method, the problem that the feature information is easy to lose in the traditional image defogging process is effectively relieved, so that the accuracy of the defogging effect and the detail reduction capability are remarkably improved. A multi-scale feature fusion mechanism is introduced, and image semantic and texture information of different scales is cooperatively integrated. Compared with a single-scale feature dependence scheme, the method has the advantage that the information expression integrity and the detail retention capability are remarkably improved. In addition, the completeness and robustness of feature representation are further enhanced through the synergistic effect of the modules, so that the model shows higher performance and stability in dealing with a defogging task in a complex scene.

The invention relates to the field of computer vision and artificial intelligence, in particular to an image event multi-mode semantic segmentation method, device and equipment, which can be applied to scenes such as automatic driving, robot perception and intelligent traffic. Time slices are divided through a fixed time window, event information is accumulated, and asynchronous event streams are converted into T * H * W voxel tensors; a selective state scanning mechanism of a Mama framework is used for replacing a self-attention mechanism of a traditional Transform, the calculation complexity is reduced while modeling global feature dependence is achieved, and the problems of video memory and delay in a Transform high-resolution scene are solved; besides, image textures and event edges are aligned through cross-space interaction, an event dynamic time sequence is captured through cross-time interaction, and modal inherent characteristics are retained through residual connection, so that feature degradation caused by excessive fusion is effectively avoided; finally, the image segmentation precision is effectively improved, and the processing efficiency and the model robustness are improved at the same time.

The embodiment of the invention discloses a multi-mode battery detection method and device, electronic equipment and a storage medium, and relates to the technical field of battery management, and the method comprises the steps: obtaining multi-dimensional original data such as battery temperature and voltage, carrying out the preprocessing, dividing the data according to a working state mode through KMeans clustering, constructing a data set through combining diversity sampling, and carrying out the detection of a multi-mode battery; and carrying out standardization processing and constructing a sliding window to obtain a time series data set. A single-layer LSTM network and a graph neural network are adopted to construct a dual-path encoder joint model based on a variational auto-encoder framework, and training is carried out. After to-be-detected data is obtained, the data is segmented by a sliding window, and the dependency relationship between the time sequence and the features is extracted respectively for splicing and compression. And restoring the sequence into a reconstruction sequence through a three-layer full-connection network, and calculating a mean square error and a regularization term to obtain a reconstruction error. And setting a dynamic threshold value, judging that an error is abnormal if the error exceeds the threshold value, calculating a feature contribution degree and generating a visual report. According to the method, the problems of low accuracy and efficiency and poor result interpretability in the prior art are effectively solved.

The application discloses a process scheduling method and a computer storage medium. The method comprises the following steps: obtaining target interaction feature data, data dependency feature data and running context feature data of a target process, respectively determining characteristic dependency strength indexes of the target interaction feature data, the data dependency feature data and the running context feature data on the target process; determining a comprehensive dependency strength index corresponding to the target process according to the characteristic dependency strength indexes corresponding to the target interaction feature data, the data dependency feature data and the running context feature data; obtaining a scheduling overhead ratio of the target process, and adjusting a process scheduling parameter of the target process according to the scheduling overhead ratio and the comprehensive dependency strength index, thereby solving the problems of high scheduling overhead, low resource utilization and limited optimization effect in the related art scheduling method, improving the resource utilization and response speed, and significantly improving the overall performance and stability of the system.

The invention provides a battery fault diagnosis method and system based on a hybrid neural network model, and belongs to the field of new energy automobile power battery fault diagnosis. According to the method, a CNN-LSTM hybrid neural network model of deep learning is adopted, and firstly, feature extraction and feature enhancement are performed through a convolution block and a residual attention module; then, time sequence modeling is carried out through a long-short-term memory network; and finally, global feature modeling is carried out through a Transform neural network algorithm, a global feature dependency relationship is established through a multi-head self-attention mechanism, and finally synchronous identification and accurate positioning of four typical faults of self-discharge abnormity, connection abnormity, sampling abnormity and insulation failure are realized.

The application belongs to the technical field of vehicle information anonymization detection, and particularly relates to a multi-type image anonymization annotation dataset construction and target coverage rate determination method. The method is based on a face and license plate image dataset with double annotation of theoretical anonymization region and anonymization features, introduces an anonymization feature extraction branch in an improved YOLOv5-L model and performs cross-modal feature fusion, combines a plurality of loss functions with theoretical anonymization region positioning loss as the core and a two-stage progressive training and difficult example mining mechanism, and realizes precise learning of the anonymization features. After normalizing the input anonymization image, the model inference obtains the theoretical anonymization region coordinates and maps them back to the original size, and through non-maximum suppression and matching of the IoU threshold, the region coverage rate is calculated to determine the positive detection, missed detection and statistical false detection rate. The application effectively overcomes the feature dependency failure and model robustness problem, and realizes high-precision, low-misjudgment anonymization detection and evaluation under various anonymization conditions.

The invention provides a surface crack laser ultrasonic quantitative detection method, device, equipment and medium, and relates to the technical field of nondestructive detection.The method comprises the steps that laser ultrasonic B-scanning data are obtained through synchronous scanning, a multi-modal characteristic wave propagation mechanism is analyzed, a flight time analysis model is established, and the flight time analysis model is analyzed; and a theoretical and experimental multi-modal characteristic mode is constructed, and discretization similarity matching is carried out, so that the depth and the inclination angle of the surface crack are stably inverted in a full-parameter space, and quantitative detection is realized. According to the method, instability caused by dependence of a single mode or a single characteristic can be avoided, synchronous and high-precision quantitative detection on the surface crack depth and the crack inclination angle under the condition of complex multi-mode wave aliasing is realized, and the problems of insufficient characterization accuracy and robustness of geometric parameters of cracks in the prior art are solved.

The application provides a disease-specific emergency whole-process auxiliary decision-making method based on multi-modal perception and AI, relates to the technical field of medical emergency, and comprises the following steps: acquiring multi-modal data of an emergency scene, constructing a feature dependency graph, and identifying disease types and symptom characteristics; indexing disease diagnosis and treatment data based on a symptom characteristic vector, determining reference cases and treatment paths; combining clinical norms to construct an action space, applying Monte Carlo search to predict expected returns, and generating an optimal treatment plan, which can improve the accuracy of emergency disease identification, shorten the treatment decision-making time, and improve the emergency intervention effect.

The invention discloses an open source software supply chain security risk intelligent assessment method, system and device based on deep learning and a medium, and belongs to the technical field of software security, and the method comprises the steps: obtaining feature data of an open source software component, and extracting software prior features, dependency relationship information and component identification information; software feature vector representation is obtained through embedded coding processing; identifying a node information missing and error dependency relationship, and carrying out noise reduction processing to obtain noise mark information; performing robust enhancement processing based on a graph neural network to obtain an enhanced dependency graph; inputting the software feature vector representation and the enhanced dependency graph into a semantic vector model to obtain fusion risk features; and a safety risk assessment result is obtained through analysis of the language model adjusted and optimized by the low-rank adapter. According to the method, the problem of noise interference in the dependency relationship is solved, effective compensation for incomplete or wrong dependency information is realized, and the accuracy and the reliability of open source software supply chain security risk assessment are improved.

The invention provides a three-dimensional human body posture estimation method based on multi-scale geometric constraint of graph convolution hybrid MLP, and relates to the technical field of posture estimation. The method comprises the following steps: acquiring a two-dimensional human joint coordinate sequence; mapping the two-dimensional human joint coordinates to a high-dimensional feature space through an embedded layer to obtain initial high-dimensional features; inputting the initial high-dimensional features into a lightweight spatial attention module, and performing adaptive weighting on the joint features to obtain attention enhancement features; the attention enhancement features are input into a multi-layer stacked MLP hybrid module, spatial topological relation modeling and channel feature dependence modeling are achieved through alternate processing of spatial paths and channel paths, and fusion features are output; performing training optimization on the model based on a multi-scale geometric constraint loss function; and decoding the fused features into three-dimensional human body joint coordinates through an output layer. According to the method, the problems that space structure modeling is insufficient and skeleton constraint is insufficient when an existing method is used for processing complex postures are solved.

The invention provides a rehabilitation training data processing method and a nervous system disease rehabilitation training system, and the method comprises the steps: carrying out the differentiation processing of continuous feature data and discrete feature data in to-be-processed rehabilitation training data, and obtaining sample data; constructing an interaction path and a retention path of the sample data based on a feature dependency relationship between the continuous feature data and the discrete feature data, and identifying high-order semantic features of each training parameter in the rehabilitation training data according to the interaction path and the retention path; determining the feature coupling degree of each high-order semantic feature and a target parameter sample in the rehabilitation training process of the target patient; and determining index keys of sensitive training parameters in the rehabilitation training data according to all the feature coupling degrees in combination with the binary masks of the training parameters, and indexing the sensitive training data from the rehabilitation training data according to the index keys. According to the technical scheme provided by the invention, the sensitive training parameters can be accurately identified in the high-dimensional heterogeneous rehabilitation training data.

The application discloses a kind of multi-modal data quality evaluation method and system of fusion meta-model and big model, first construct the evaluation meta-model including object set, morphism set, combination sub-set and feature dependency graph, then receive multi-modal data evaluation task and automatically convert multi-modal data evaluation task into a composite morphism in evaluation meta-model;The application realizes the function of multi-modal data quality evaluation with the object set, morphism set, combination sub-set and feature dependency graph for constructing the evaluation meta-model based on category theory, and by abstracting evaluation operation as morphism, evaluation process as composite morphism and shared computing as dependency graph, a combinable, optimizable and reusable evaluation framework can be constructed, and by evaluation functor, heterogeneous score output can be unified standardized, the application can unify the logic of multi-modal data quality evaluation, automatic scheduling and knowledge sedimentation.

The application provides a network security malicious code binary search method and system, and relates to the technical field of data processing.The method comprises the following steps: based on a function level control flow semantic feature vector, the multi-dimensional structural similarity between a to-be-detected binary file and each standard binary reference file is calculated; according to the multi-dimensional structural similarity, the final matched standard binary reference file is determined; the to-be-detected binary file and the final matched standard binary reference file are subjected to fine-grained difference analysis, and a fine-grained difference analysis result is obtained; based on the fine-grained difference analysis result, combined with sensitive data operation behavior feature detection, it is judged whether the to-be-detected binary file is a malicious tampered version, and a malicious code injection point is located.The application overcomes the defects that the prior art is too dependent on fixed features and sensitive to compilation optimization.

The application discloses a kind of nuclear instance segmentation model and method based on attention and ellipse regularization, belong to deep learning image processing technical field, model includes from shallow to deep window attention mechanism feature extraction module, long-distance feature dependent attention fusion module and ellipse regularization module, input image is extracted feature by from shallow to deep window attention mechanism feature extraction module, then by long-distance feature dependent attention feature fusion, finally by ellipse regularization module constraint model.The application uses the above instance segmentation model and method, for the small and elliptical shape of cell nucleus is constrained training, strengthens the image feature extraction capability based on bounding box instance segmentation, more effectively separates the adherent cell nucleus, can extract more complete cell nucleus density, morphology and location information.

The application relates to the technical field of structural health monitoring and engineering safety, in particular to a dam automatic multi-element data fusion monitoring system and method based on the Internet of Things; the method comprises the following steps: an initial state reference without historical sample dependence is established by deploying checked heterogeneous sensing units on site and constructing a dynamic baseline mechanism; then, cross-scale mapping matrices and feature-dependent correction coefficients are used to perform multi-domain analysis and scale self-adaptive adjustment on time series data, so that the difference capturing capability is enhanced; a dynamic credibility adjustment and multi-source weighted fusion method is adopted at the edge node, preliminary event discrimination is performed in combination with physical connectivity and neighborhood cooperation; finally, abnormal diffusion trend analysis and dynamic early warning level determination are realized based on a space-time evolution matrix and a probability prediction model, and a response strategy and system parameter optimization are automatically generated through a closed-loop feedback mechanism. The application significantly improves the accuracy, self-adaptive capability and risk control level of dam monitoring.

The invention discloses a ground penetrating radar underground cavity intelligent detection model, method and system based on dynamic characteristic calibration, and the model comprises an encoder, a decoder and a dynamic characteristic calibration module. The dynamic feature calibration module is arranged in a jump connection feature transmission path between the encoder and the decoder, and is used for modeling feature dependency relationships between channels and space through a channel attention branch and a space attention branch respectively, and carrying out dynamic calibration on features extracted by the encoder; and the calibrated features are transmitted to a decoder for feature fusion. The detection model disclosed by the invention has the following advantages: (1) high detection precision: the dynamic feature calibration module can adaptively focus on the features of the cavity target and effectively suppress background clutter and noise, so that the model is significantly superior to the existing mainstream model in key indexes such as average Dice coefficient, recall rate and the like; (2) the small target detection capability is strong: the attention and segmentation capability on small-scale holes and weak echo signals is enhanced, and the small target IoU index is ahead of the comparison model;

This invention discloses a dynamic scattering medium imaging method based on Mamba channel feature dependency and dual-domain learning, belonging to the field of scattering medium imaging technology. It employs a ScatMamba network with an encoder-decoder architecture to extract initial features from speckle images formed by dynamic scattering media. Multi-scale high-level semantic features are extracted through downsampling and channel filtering enhancement of the Mamba module, and frequency domain global feature modulation is achieved using a spectral modulation module at the encoder bottleneck layer. The decoder fuses multi-scale features through upsampling and skip connections, outputting a clear reconstructed image. End-to-end training is completed using a spatial-frequency dual-domain composite loss function. This invention achieves joint modeling of channel feature dependency and spatial long-range dependency through channel filtering and enhanced Mamba layers, utilizes a spectral modulation layer for adaptive control of high and low frequency components, and strengthens the robustness of the reconstruction mapping through a dual-domain loss function. This enables efficient decoupling of speckle features in dynamic scattering environments, restoring the global structure and fine texture of the target.

The invention relates to the technical field of computer network security detection, in particular to a computer network security detection system and method. According to the method, firstly, the feature dependency degree related to the network attack is calculated through the feature dependency function, the key features are screened out according to the feature dependency degree, the feature dimension of the network flow data is greatly reduced, and therefore the data processing efficiency is improved. Secondly, the invention provides a KNN-LightGBM joint voting model, and the model is applied to the first-stage detection of a network security detection system, and the model can quickly identify and screen out potential high-risk data, thereby effectively reducing the workload of subsequent detection tasks, and improving the overall detection efficiency. And finally, the natural language processing model BERT is introduced into the second-stage detection of the network security detection system, and the network traffic data is converted into a text sequence form for analysis, so that the detection precision of complex network attacks is enhanced, and accurate attack type classification is realized.

Computer-implemented method, comprising: Providing a service manager that interacts with a database management system (DBMS), where the DBMS is a software application designed to enable the definition, creation, querying, updating, and management of databases; Automatic and dynamic registration of multiple DMBS services with the service manager during the DBMS runtime, wherein: Each of the multiple DBMS services processes data that is stored in a database managed by the DBMS; Each of the multiple DBMS services is implemented as a singleton, where a singleton is a software object implemented in such a way that only one instance of a specific type of software object can be instantiated at a time; and A first service among several DBMS services allows the service manager to check whether a cached singleton instance has already been initialized; and Managing dependencies between the multiple registered DBMS services by the service administrator, wherein: The dependencies are managed based on the source code of the DBMS service, which specifies dependency services required by the DBMS service; The class name of each dependency service is automatically converted into a string that is used by the service manager as an argument for managing the service; The dependencies of the services are represented in a single data structure and are identified and updated during the runtime of the DBMS using a deep search in a dependency tree that represents dependencies of the DBMS services registered with the service manager; the global variable represents a specific service that is to be registered with and managed by the service administrator; and Managing dependencies also features: in response to the creation of the global variable, execution of a constructor of a type associated with the global variable; and Identifying cyclic dependencies based on a representation of the dependencies of the DBMS services in the form of a directed acyclic graph.

The invention relates to the technical field of deep space exploration, and provides a deep space task evaluation method and device based on data fusion and multi-task attention. According to the method, intelligent evaluation of various deep space exploration tasks is realized by combining a dynamic multi-granularity time window fusion mechanism with a multi-task attention neural network model. Wherein the dynamic multi-granularity time window fusion mechanism can effectively integrate data streams with different sampling frequencies and updating periods, and differentiated processing strategies are adopted for data sources with different frequencies, so that high-quality input is provided for downstream deep learning tasks. The multi-task attention neural network model can dynamically capture complex feature dependency relationships in heterogeneous data and distribute larger weights for important information related to tasks, so that the perception accuracy and robustness of the model are improved, and the multi-task parallel modeling and the result joint optimization can improve the perception accuracy and robustness of the model. The real-time performance and effectiveness of task response can be improved, and the failure risk caused by incomplete data or task isolation is reduced.

The application discloses a kind of high-efficiency semantic segmentation networks, suitable for medical image analysis, automatic driving etc. The network adopts encoder-decoder architecture, combines CNN with Transformer (converter model), balances global modeling and computational efficiency. Encoder extracts multi-scale features through lightweight convolution, and introduces spatial selection module: its gate convolution splits channel into gate signal and reserved information, and key spatial features are activated and strengthened by Sigmoid activation function; Grouped pooling module extracts details using multi-scale pooling, and restores channels after upsampling and splicing. The decoder fuses multi-scale features and restores resolution through upsampling, enhances efficient channel attention, fuses global max pooling and average pooling, generates channel weight using one-dimensional convolution, and optimizes feature dependence. The design improves small target segmentation accuracy through gate mechanism and multi-scale pooling, is lightweight and easy to expand, and has high precision and practicality.

The invention discloses an arrhythmia detection model modeling method based on multi-scale spatial-temporal feature fusion, and the method comprises the steps: designing a multi-scale selection fusion embedding module, and achieving the efficient fusion of multi-scale time features in an ECG signal; an arrhythmia detection model is designed, a space-time agency attention module is introduced into one branch, space-time features of ECG signals in input leads are extracted, fused and enhanced, and in the other branch, the cosine similarity of approximation coefficients between all lead pairs is calculated and a threshold value is applied, so that the time-space characteristics of the ECG signals in the leads are extracted, fused and enhanced. Constructing a graph structure of the correlation between the leads to quantify a feature dependency relationship between the leads, and performing operation depth extraction on spatial topological features between the leads through a multi-layer graph convolutional network module; in-lead multi-scale spatial-temporal features and inter-lead spatial features obtained by two branches of the arrhythmia detection model are fused cooperatively, so that the network can understand ECG signals more comprehensively, and obtained mixed features are input into a full-connection layer for prediction and classification.