196results about How to "Enhanced Representational Capabilities" patented technology

The invention discloses a multi-modal remote sensing small target recognition method based on common-differential mode collaborative interactive fusion, and relates to the technical field of image recognition. Performing feature extraction and interactive fusion based on a double-flow backbone feature extraction network in the multi-modal remote sensing small target recognition model to obtain an infrared enhanced fusion feature and a visible light enhanced fusion feature respectively; the infrared enhancement fusion features and the visible light enhancement fusion features output by the corresponding convolution down-sampling module and the output layer in the two backbone feature extraction channels are spliced and then are subjected to neck feature fusion output to obtain a prediction image with a target recognition result. According to the method, infrared and visible light modal features are fully utilized, and common features and difference features of the infrared and visible light modal features are subjected to collaborative modeling and interactive fusion, so that the characterization level of the small target is effectively improved, and recognition of the remote sensing small target is more robust.

The invention discloses an action sports scoring method and device based on big data analysis, and relates to the technical field of computer vision, and the method comprises the following steps: S1, constructing a human body posture tensor manifold; s2, generating aligned action track characteristics; s3, generating symmetrical positive definite manifold features; s4, generating deep manifold distribution parameters; s5, quantifying global distribution deviation characteristics; s6, calculating a residual vector based on the deep manifold distribution parameters and the standard action distribution parameters, inputting the residual vector into the improved AGCN model for processing, constructing an adaptive topology based on the residual vector, and performing aggregation analysis to obtain a joint-level physical angle error after multi-scale space-time convolution and manifold enhancement; and S7, outputting a comprehensive score vector. According to the method, the limitation that a traditional method only depends on local geometric features and ignores global topological structure constraints and statistical distribution priori is overcome, and an efficient solution is provided for intelligent scoring of sports actions.

The invention discloses a drilling pump pressure prediction method based on an artificial neural network, and the method comprises the following steps: S1, data collection; s2, data preprocessing; s3, constructing a neural network model; s4, model training and optimization; and S5, model verification and deployment. According to the method, multi-channel data fusion and time synchronization optimization are innovatively adopted, so that the data quality and consistency are improved; in combination with a deep neural network and a self-adaptive optimization strategy, the precision and generalization ability of pump pressure prediction of the drilling pump are improved; and an online updating mechanism is introduced, so that the model can be dynamically optimized according to real-time data, the defects of low prediction precision, poor adaptability and difficulty in real-time updating of a traditional method are overcome, and an efficient and reliable prediction means is provided for intelligent drilling control.

The invention relates to the technical field of industrial big data, in particular to an Internet of Things sensing and data monitoring integrated system, which comprises an Internet of Things sensing module used for acquiring a sensing parameter set of a distributed simulation unit of an intelligent power plant and a historical performance evolution trend reflecting a dynamic evolution rule; the twin mapping module is used for realizing signal space-time synchronization by utilizing a time service protocol, calling a multi-physics field evolution model to calculate thermal stress offset so as to execute numerical compensation, and generating a high-fidelity physical feature vector; the collaborative decision-making module shares observation information by using a multi-agent distributed collaborative architecture, identifies an electromagnetic-dynamic coupling response mode through a joint reward function and outputs a collaborative optimization strategy; and the execution module is used for generating an evaluation result by calculating the numerical deviation between the physical feature vector and the design envelope, and adjusting the excitation control variable according to a strategy until the system converges to a target equilibrium value. According to the method, the steady state monitoring of the power system is realized through twin mapping and collaborative decision.

The application provides a ConvNeXt model SAR ship classification method based on multi-level feature cooperative interaction. In order to effectively utilize traditional manual features, the method uses a Canny-based edge detection method, captures edge information of objects in an image through multi-level filtering and edge gradient analysis, and inputs the result into a ConvNeXt model for feature extraction. In addition, a multi-level branch structure is expanded in the ConvNeXt model, which is used for deep extraction of feature information at different scales, so as to capture various information of the image at low-level features and high-level features. Finally, cross fusion with traditional manual features is carried out in the multi-scale feature extraction process, so as to improve the classification accuracy of the model.

The application discloses an infrared image generation method and system based on a physical prior constraint generative adversarial network, relates to the field of infrared image generation, and aims to solve the problems of insufficient physical reality and lack of representation ability of infrared radiation characteristics of the existing infrared image generation method. The application first designs a physical prior knowledge learning network, which is used to learn the prior representation related to the infrared radiation characteristics from the infrared image, and realizes the prediction of the infrared radiation related parameters through a deep learning network. Secondly, a generative adversarial network based on the physical prior constraint is designed based on the proposed physical prior knowledge learning network, the physical prior features are introduced into the image generation process, and the learning ability of the generator to the overall distribution of the infrared radiation and the representation ability of the generator to the infrared radiation characteristics are enhanced. The method breaks through the limitation that the traditional generator only relies on shallow texture mapping. The application is suitable for the fields of visible light to infrared image generation, infrared vision algorithm research and development.

The application relates to the field of sanitation operation scheduling, in particular to a sanitation operation scheduling method based on the Internet of Things, which comprises the following steps: determining an iterative stop condition of a load CEEMDAN model based on time local weighted variance of load data, so as to generate vehicle load characteristics; generating vehicle oil consumption characteristics through an oil consumption CEEMDAN model based on statistical values of load data residuals and oil consumption data residuals, so as to calculate oil consumption channel residual signals; generating comprehensive vehicle carrying characteristics through a feature extraction model based on a multilayer perception machine, a gating mechanism and a cross-modal attention mechanism; and generating corrected vehicle garbage load through a long short-term memory network and a fully connected layer architecture. The application realizes efficient noise reduction and accurate decomposition of non-stationary time series data of oil consumption and load, realizes accurate processing of Internet of Things data of garbage transfer vehicles, improves the operation efficiency of the garbage transfer vehicles and reduces the operation cost.

The present application relates to the technical field of gait analysis, and discloses a gait plantar pressure intelligent prediction system based on a Transformer structure, which comprises a gait collection platform, a plantar pressure collection device, a gait kinematics collection device, a central processing server and a display terminal. The system obtains initial plantar pressure distribution through a lower limb multi-rigid-body chain biomechanical model and foot-ground contact mechanics solution; and constructs a gated prefix memory sliding window Transformer in the central processing server, performs time series modeling and spatial correction on the initial pressure matrix and gait characteristics, and generates plantar pressure prediction results conforming to the biomechanical law. The system effectively solves the problems of unstable prediction, weak generalization ability and lack of interpretability of traditional methods, realizes higher-precision plantar pressure prediction, and can be used for gait evaluation, rehabilitation training and orthosis customization.

The application claims a small sample bearing fault diagnosis method based on Mamba and double granularity cross-domain alignment, which is used to solve the problem of low diagnosis accuracy caused by large distribution difference of bearing monitoring data across working conditions and scarcity of target working condition samples. The method constructs a double branch space-channel feature extraction network CA-Mamba, adopts an improved Mamba module and channel attention to enhance key response, and fuses deep features through a selective bridging module. Then, a domain discriminator and a group discriminator are constructed to realize cross-domain feature confusion and class boundary constraint; relying on the confidence weighted pseudo label PSMMD and dynamic memory enhanced contrast learning DMECL, the target domain feature is constrained from the class-instance double granularity, so that the target domain sample is aligned to the homologous source domain prototype, and finally the covariance measurement is used to complete the target domain classification. The application can extract domain-invariant and high-discriminative features under small sample conditions, and improve the diagnosis accuracy under cross-domain conditions.

The application provides a SAR target detection method combining feature fusion and position relationship awareness, comprising the following steps: constructing a real-time target detection model based on a Transformer, introducing a spatial detail compensation path into a mixed encoder and containing a mixed representation enhancement module; a Transformer decoder comprises a position relationship awareness module; inputting a SAR image into a trained target detection network, inputting multi-scale fusion features into the Transformer decoder, adjusting self-attention weight distribution according to the position relationship between queries in the same layer by using the position relationship awareness module, guiding spatial layout reasoning, and outputting a target detection result; and the target detection network is trained by using a weighted intersection over union zoom loss function based on a dynamic IoU weight guiding mechanism as a classification loss function. In this way, the target detection rate, positioning accuracy and overall robustness in a small target and dense scene are improved, and feature confusion under strong scattering interference is alleviated.

This invention discloses a smart contract vulnerability detection model and device. The model includes a code structure feature extraction module, a thought chain text feature extraction module, a semantic space adaptation module, a multi-head cross-attention module, and a vulnerability classification module. Based on code structure information modeling, this invention further introduces vulnerability reasoning thought chain text features and achieves effective alignment between code structure semantics and vulnerability reasoning semantics through the synergistic effect of semantic space adaptation and multi-head cross-attention. Compared to detection methods that rely solely on overall semantic modeling of a single code modality, this invention no longer depends solely on the overall statistical features of the source code for vulnerability identification. Instead, guided by vulnerability reasoning semantics, it further focuses on core code segments strongly related to the vulnerability triggering logic, reducing the interference of irrelevant business code and redundant code on the feature extraction process, and improving the accuracy and robustness of vulnerability detection.

This invention discloses a method, apparatus, and device for predicting the activity of chemically modified siRNA, comprising: acquiring the original base sequence and chemical modification information of the target siRNA, and retrieving the physicochemical properties of the target siRNA; encoding the original base sequence, chemical modification information, and physicochemical properties as features; and generating a prediction result of the silencing efficiency of the target siRNA based on the encoded features using a pre-constructed prediction model; wherein the pre-constructed prediction model includes: a feature fusion sub-model and a classification sub-model; the feature fusion sub-model is used to perform feature fusion on the encoded features based on a cross-attention mechanism; and the classification sub-model is used to generate a prediction result of the silencing efficiency based on the fused features. This method, based on a multi-dimensional, multi-view learning strategy and an attention mechanism fusion model, significantly improves the algorithm's ability to represent chemically modified siRNA data, and, combined with the nonlinear data fitting ability of the deep learning framework, improves the accuracy of the algorithm in predicting the drug activity of chemically modified siRNA.

The application discloses a large field of view structure image classification method based on continuous deformation modeling, mainly solves the problem that the existing dynamic deformation modeling is insufficient in integrity, resulting in poor classification effect. Including: 1) constructing progressive structure similar samples, realizing self-learning adaptation of non-rigid structure intra-class change; 2) constructing a multi-branch attention representation learning network composed of three parallel branches of double-flow spatial attention, adaptive fusion channel attention and original features; 3) configuring a multi-branch selection fusion module based on the learning network, and constructing a continuous deformation feature extraction network; 4) inputting the continuous deformation fusion features output by the extraction network into a classifier to realize self-learning classification of aurora large field of view structure images. The application can automatically learn the structure change rule under the condition of limited labeling or unlabeled, significantly improve the discrimination ability and generalization performance of complex deformation structure, and is suitable for aurora, cloud cluster and other non-rigid deformation image analysis.

This invention proposes a mechanism-aware synergistic anticancer drug combination prediction method and system, belonging to the field of biomedical technology. It includes: extracting node-level feature representations of the first and second drugs through a drug encoder, and extracting molecular representations of the cell line through a cell line encoder; generating an attention tensor based on a trilinear cross-modal synergistic strategy using a trilinear attention network to explicitly capture the synergistic dependencies of the drug combination in a specific cellular environment, and updating and enhancing the features through context fusion; finally, concatenating the enhanced features and outputting a synergistic score via a multilayer perceptron. This invention can directly model the joint interaction relationship between the first drug, the second drug, and the cell line, thereby accurately predicting the synergistic effect of the anticancer drug combination.

The invention discloses a multi-label smell description prediction method, and relates to the field of compound smell prediction, and the method comprises the steps: obtaining compound identification information, molecular structure descriptors and smell label data, and constructing a multi-label smell data set; generating a molecular structure feature vector through a molecular fingerprint coding technology, and extracting a multi-dimensional descriptor reflecting the physicochemical properties of molecules; compressing the molecular fingerprint features to a low-dimensional space through a dimension reduction algorithm; performing unbalanced data processing on the training set, fusing the dimension-reduced molecular fingerprints with the molecular descriptors to form a joint feature matrix, and configuring a class weight balance mechanism and overfitting suppression parameters by adopting a multi-label classification architecture; independently optimizing a probability threshold for each odor label based on the verification set; and outputting a multi-odor label combination prediction result according to the target molecule identification information. According to the scheme, the multi-odor characteristics of the compound can be accurately depicted, and the combined recognition accuracy of the compound odor is remarkably improved.

This invention provides a multimodal fusion-based method for correlation analysis of deep-sea environmental data outside the Yangtze River Estuary, belonging to the field of deep-sea environmental analysis technology. This invention achieves accurate data registration by establishing a spatiotemporal standardization model for multi-source data and employing an adaptive spatiotemporal kriging interpolation algorithm. It utilizes a sparse coding marine signal separation algorithm to construct an overcomplete dictionary to separate mixed signals and extract pure features. Based on an attention mechanism, a feature extraction network is constructed to automatically learn deep feature representations of physical, chemical, and biological parameters. A matrix rank deficiency detection algorithm is used to automatically supplement feature compensation vectors to ensure feature integrity. A cross-scale attention mechanism based on wavelet transform is constructed to fuse multi-scale features. A Shapley value interpretability evaluation system is established to quantify the importance of correlated features. This invention solves the technical problem of inaccurate correlated feature extraction during the fusion of multi-source heterogeneous marine environmental data.

The application discloses a kind of based on big language model's automobile data classification grading system, comprising: data object generation module, for constructing data object;Incremental vocabulary reconstruction module, for expanding vocabulary;Uniform coding module, for generating input representation;Rotary segment offset position writing module, for generating coding result;Fastformer feature extraction module, for extracting hierarchical representation;Intermediate layer distillation training module, for updating Fastformer parameter;Result output module, for outputting automobile data category, sensitive attribute label and data level.The application realizes the efficient processing and accurate output of automobile data classification grading, improves professional semantic modeling capability and practical application value.

The application belongs to the field of artificial intelligence model compression and edge deployment in intelligent operation and maintenance and fault diagnosis of power equipment, and discloses a transformer cooperative distillation incremental learning method and system based on time sequence consistency, which comprises the following steps: a teacher model is used to screen a no-label sample set to obtain a pseudo-label sample set, the pseudo-label sample set is combined with an original label sample set to obtain a distillation training sample set; a multi-mechanism cooperative distillation training scheme containing soft and hard label joint distillation, time sequence consistency distillation and multi-task distillation is constructed, and a student model is subjected to distillation training through the multi-mechanism cooperative distillation training scheme; and a cooperative distillation incremental learning scheme in which a cloud end continuously updates a teacher model to adapt to new multi-source monitoring data and a student model after edge distillation generates a pseudo-label to expand the distillation training sample set is constructed. The application can be used for realizing high-precision and low-cost training and continuous updating of a student model in a resource-limited terminal or an online scene.

The invention discloses a CBN grinding wheel non-contact type grinding high-temperature alloy taper hole abrasion prediction method, and relates to the technical field of state monitoring and intelligent prediction in the precise grinding machining process. The method mainly comprises the following steps: calibrating a dynamic contact arc acoustic propagation model by using a system identification method to obtain dynamic geometric parameters, and constructing acoustic-vibration geometric coupling characteristics in combination with multi-mode signals in a grinding process; correcting the acoustic-vibration geometric coupling features by using a maximum mean difference algorithm and an adaptive window algorithm, training a wear prediction model by using the corrected acoustic-vibration geometric coupling features, and performing regression analysis on the corrected acoustic-vibration geometric coupling features by using the trained wear prediction model to obtain a prediction result. According to the grinding wheel abrasion prediction method for high-temperature alloy taper hole grinding, the adaptability, robustness and sensitivity of abrasion state characterization under complex variable working conditions can be improved.

The application discloses an optical remote sensing image semantic segmentation method and relates to the technical field of optical remote sensing image processing. The method is composed of a progressive context aggregation mechanism and a dynamic wavelet transform attention module, query-key-value features are generated in a hierarchical manner, and progressive fusion of low-level spatial details and high-level semantic information is realized; and grouping multi-level wavelet decomposition is adopted to adaptively and weightedly enhance low-frequency semantic features and high-frequency detail features, and dynamic frequency filtering is realized in combination with query guidance. Compared with the prior art, the application can effectively capture multi-scale features and fine spatial details of remote sensing images, fully model global context and long-distance dependency, and is compatible with mainstream encoders, so that more accurate and robust semantic segmentation can be realized on public data sets.

The application discloses a self-supervised device running abnormal sound detection method, characterized in that single-channel running normal sound signals of multiple different types of devices in a running state are collected, and frequency spectrum graphs and time spectrum graphs are obtained by conversion; the frequency spectrum graphs and the time spectrum graphs are spliced to obtain double-channel time-frequency feature graphs, the double-channel time-frequency feature graphs after feature enhancement are input into a dynamic feature extraction module, so that final one-dimensional audio feature vectors of different types of devices are extracted, and a probability density function of the final one-dimensional audio feature vectors is fitted; single-channel running sound signals of a device to be detected in a running state are collected, and the above steps are repeated to obtain the probability density of one-dimensional audio feature vectors of the device to be detected; finally, the probability density is input into a classifier to complete the discrimination of normal and abnormal sounds; the method can learn effective features of audio of different types of devices through time-frequency fusion and dynamic feature extraction, and the accuracy and reliability of abnormal sound detection are improved.

The application discloses an intelligent conference management system and method based on a large model, relates to the technical field of intelligent conference management, and comprises the following steps: extracting and processing a project sequence of multiple conferences to obtain an enhanced representation of a current conference; recognizing audio data of the current conference to obtain a text segment sequence; projecting the text segment sequence and the enhanced conference representation to obtain a context-aware segment representation and calculate a dynamic importance weight to obtain a first candidate decision point set; iteratively compressing the first candidate decision point set through a double-constraint strategy to obtain a second candidate decision point set; for each decision point in the second candidate decision point set, extracting a feature vector of the decision point from multi-modal data of the current conference and projecting the feature vector into a projection vector; calculating a cosine similarity between the projection vectors to obtain a final confidence; and associating the final confidence with the corresponding decision point to obtain a third candidate decision point set.

The invention discloses a rock slice image classification method based on domain self-adaption, and the method comprises the steps: collecting rock slice image data, the method comprises the following steps: acquiring a rock slice image, performing multi-scale data preprocessing and geological field data enhancement, extracting multi-level visual features of the rock slice image by utilizing a pre-trained DINOv3 model, performing field specialized adaptation on general visual features through a geological field adaptive Adapter module, and enhancing rock slice discriminative feature representation by adopting a double-path attention mechanism. Constructing a progressive hierarchical classification head to realize coarse-to-fine rock classification; and designing a multi-stage progressive training strategy to optimize the overall performance of the model. According to the method, mineral composition and structural features of the rock slices under different scales can be accurately captured, and multi-scale features and an attention mechanism are fully utilized, so that accurate classification of the rock slices is realized, and the accuracy and reliability of rock slice identification are remarkably improved.

This application discloses an auxiliary classification system and method based on multi-view spatiotemporal interaction and difference compensation, relating to the field of auxiliary classification. The method includes: acquiring time-series data from multiple brain regions and calculating Pearson correlation matrices and partial correlation matrices; constructing first and second-dimensional time matrices based on features extracted from the time-series data; generating multiple spatial connectivity matrices based on the Pearson correlation matrix and partial correlation matrix, thereby constructing first and second-dimensional spatial matrices; performing bidirectional cross-attention interaction on the time and spatial matrices of the two dimensions respectively to obtain corresponding graph structure representations and node representations; generating compensation terms based on the differences between the graph structure representations and node representations of the two dimensions; enhancing the graph structure representation and node representation of the second dimension using the compensation terms to obtain fused features; and inputting the fused features into a graph convolutional network to output classification results. This invention improves the accuracy of auxiliary classification through two-dimensional spatiotemporal interaction and difference compensation.

This invention relates to the field of smart agriculture information technology, and discloses an agricultural pest and disease prediction system and method based on multimodal data fusion and edge AI, comprising: a data acquisition module, an edge AI analysis module, and a communication interaction module; the data acquisition module is used to collect agricultural monitoring data containing agricultural environmental information and crop status information; the edge AI analysis module is used to call a pest and disease prediction model that integrates multimodal agricultural feature extraction to perform multimodal joint feature extraction on the agricultural monitoring data containing agricultural environmental information and crop status information, and infer pest and disease prediction results based on the multimodal joint features, and generate pest and disease early warning information based on the pest and disease prediction results; the communication interaction module is used to upload the agricultural monitoring data collected by the data acquisition module and the pest and disease prediction results and pest and disease early warning information output by the edge AI analysis module to a cloud platform. This invention enables agricultural pest and disease prediction on resource-constrained edge devices.

The application discloses a Transformer oil well production prediction method and device based on gating constraints and Kalman filtering, and a medium, which comprises the following steps: first, obtaining oil well historical production data and preprocessing; then, using a mutual information method to screen input features related to target production, and constructing conditional gating constraint features representing the opening and closing states of the oil well; then, inputting the processed time sequence samples into a Transformer prediction model to obtain initial prediction results of the oil well production; then, performing state constraint on the initial prediction results according to the conditional gating constraint features, and performing causal recursive residual correction on the prediction results of subsequent time based on the current time prediction residual by using Kalman filtering to obtain final prediction results; further, performing grid search on Kalman filtering parameters Q and R for optimization, and determining the optimal parameter range by combining a heat map. The application can effectively reduce the prediction error, and improve the accuracy, stability and robustness of the oil well production prediction.

The application discloses a hyperspectral anomaly detection method based on background clustering constraint and potential feature separation, and relates to the technical field of hyperspectral remote sensing detection applications. First, band selection and partition generation of a hyperspectral original image are carried out to generate fixed-size image blocks. Then, a basic image reconstruction framework is constructed based on a lightweight spectral-spatial feature extraction encoder and a convolution decoder. Then, a pixel-level deep clustering network is introduced to complete latent feature clustering modeling, calibrate the background clustering label of the image block, and obtain the intra-class RX detection score in the feature space. Then, the intra-class RX score and the reconstructed image are reversely mapped to the original image space to construct a global RX response map and a reconstructed full image. Finally, a weight matrix is generated by using the reconstruction error between the original image and the reconstructed full image, the global RX response map is modulated, and the final anomaly detection result is output. The application fuses the background clustering constraint and the potential feature separation mechanism, effectively improves the detection accuracy of small anomaly targets, and significantly suppresses false alarm interference.

This invention discloses a real-time monitoring method for vehicle lane-crossing behavior based on image semantic segmentation, comprising the following steps: acquiring traffic video data to obtain a road image sequence and performing preprocessing; inputting the image semantic segmentation model for semantic segmentation processing to obtain a semantic segmentation result image; performing morphological processing and connected component analysis to obtain the set of pixel boundaries of the vehicle region and the set of pixel boundaries of the lane lines; calculating the spatial position data of the lane lines and performing perspective correction processing to obtain perspective-corrected lane line spatial position data; generating the position region of the vehicle region in the image coordinate system; performing target association matching to generate vehicle motion trajectory data; establishing spatial relationship data and forming relative position changes; performing temporal consistency analysis and real-time determination, and outputting the vehicle lane-crossing behavior detection result, thereby achieving continuous, accurate, and real-time monitoring of vehicle lane-crossing behavior.

The application relates to the field of battery life prediction, in particular to a multi-feature fuel cell life prediction method based on a long short-term memory network; the method comprises the following steps: collecting voltage attenuation data and collecting running state data to construct an original input data set; analyzing the running state data, calculating a correlation coefficient according to the voltage attenuation data and each running state data, analyzing the correlation coefficient, and obtaining an input feature set; processing the input feature set and the voltage attenuation data to construct a neural network training data structure; dividing the training data structure into a training set and a test set; constructing a network architecture solution space to be searched, analyzing the network architecture solution space, constructing a fitness function, and determining an optimal network architecture parameter combination based on the fitness function; and inputting the training set into the optimal network architecture parameter combination for training to generate a target detection prediction result. The application can improve the prediction accuracy of the voltage attenuation trend.

The present application relates to the technical field of Pickleball, and specifically relates to an intelligent wearable system for monitoring sports posture. The system synchronously collects multi-axis acceleration and angular velocity original signals before and after the player hits the ball through the inertial measurement unit built in the wearable device, generates time series after time window interception and normalization processing; removes noise and baseline drift by wavelet transform; fuses time domain and frequency domain features to construct a high-dimensional comprehensive feature vector; realizes preliminary classification of actions through a long short-term memory network, refines similar action types by combining a convolutional neural network, and finally outputs accurate classification results through a support vector machine; dynamically optimizes model parameters based on a real-time feedback mechanism, and enhances local feature representation through a channel attention mechanism. The system can accurately identify more than fifteen Pickleball special technical actions, has high robustness and self-adaptive ability, can provide real-time posture specification guidance for athletes, and helps to improve training efficiency and prevent sports injuries.