838 results about "Mahalanobis distance" patented technology

The invention discloses a network security defense method and system based on incremental network attack analysis learning. The method comprises the following steps: collecting initial network flow data, and extracting a feature vector; and collecting real-time network flow data, performing segmentation processing based on a sliding time window, extracting time sequence association features from the segmented data, and matching the time sequence association features with the feature library to identify potential attacks or abnormal behaviors. And when the time sequence correlation feature is not matched with the feature library, marking the time sequence correlation feature as a candidate novel attack feature, calculating a mahalanobis distance between the time sequence correlation feature and a known attack feature to determine the attack variability, and dynamically adjusting the weight of the time sequence correlation feature. And inputting the adjusted feature weight and the real-time flow feature into a deep reinforcement learning detection model, and generating and updating a network security defense strategy. The scheme of the invention can effectively identify novel attack behaviors, dynamically respond and optimize defense strategies, and improve network security.

The invention relates to a roadside signboard identification method based on unmanned aerial vehicle three-dimensional point cloud lightweight semantic segmentation, and belongs to the technical field of intelligent traffic. The method comprises the following steps: optimizing a point cloud acquisition path through adaptive flight control; adopting an RSPAE algorithm to enhance local geometric features of the point cloud; converting the point cloud into a three-channel fusion image (a depth image, an intensity image and a local depth variance image); extracting multi-scale features by using a double-branch neural network, and fusing the aligned features through a GFM module and a CFM module; the lightweight decoder recovers a high-precision semantic segmentation map; and generating a final identification result by combining geographical registration and multi-frame redundancy suppression. And state evaluation and anomaly detection are realized based on an MLP scoring device and a mahalanobis distance. According to the method, the segmentation precision, the reasoning speed and the positioning precision are remarkably improved in a complex scene, the method is suitable for deployment of embedded equipment, and the problems of low efficiency and high omission ratio in the prior art are solved.

The invention provides a bridge structure monitoring method and device based on a microwave deformation radar, and relates to the technical field of bridge structure monitoring, and the method comprises the steps: obtaining the multi-point three-dimensional displacement data of a bridge structure through the microwave deformation radar, carrying out the thermal expansion pseudo displacement compensation and multi-point space smoothing through combining with temperature information, and obtaining a displacement field after environment correction; secondly, extracting a vertical component and separating the vertical component into a static deformation component and a dynamic vibration component by adopting variational mode decomposition; further analyzing and identifying a decoupling region through a time window coherence coefficient, performing recursive quantitative analysis, bispectrum analysis and energy distribution entropy calculation on a dynamic signal of the region, and constructing a high-order damage sensitive feature set; and finally, the dynamic characteristics and the static curvature change are fused to form a comprehensive degradation degree index, the deviation degree is judged according to working condition classification and the mahalanobis distance, and multi-dimensional and cross-working-condition degradation identification and risk early warning of the bridge structure are achieved.

The invention belongs to the technical field of tunnel inspection and auxiliary rescue, and particularly discloses an unmanned aerial vehicle multi-source fusion inspection and auxiliary rescue method based on an urban tunnel, multiple sensors are integrated on an unmanned aerial vehicle platform, and multi-source data are processed through a dynamic weighting factor graph optimization method based on a sliding window. By introducing a Mahalanobis distance consistency test mechanism, dynamically evaluating the consistency between sensor observation data and IMU prediction data, screening out data with high credibility, and endowing each factor with different optimization weights, the better the quality of the sensor observation data is, the higher the factor weight is, so that the robustness of final target estimation is improved, and the target estimation accuracy is improved. And then uniformly converting sensor data at each moment into a global coordinate system by using an SLAM framework, gradually constructing a three-dimensional map of the tunnel, and finally realizing high-precision real-time positioning of the unmanned aerial vehicle and construction of the three-dimensional map of the tunnel for subsequent inspection and auxiliary rescue tasks.

The invention relates to the technical field of electromagnetic spectrum monitoring, and particularly discloses a distance-based electromagnetic spectrum monitoring abnormal data detection method, which comprises the following steps of: performing short-time Fourier transform and normalization processing on an acquired original signal to generate an energy density distribution characteristic graph; constructing a multivariate Gaussian distribution model based on non-abnormal historical data; during real-time monitoring, the mahalanobis distance between the collected data and the mean vector of the historical model is calculated after the collected data is preprocessed. And comparing the distance metric value with a preset threshold value to preliminarily judge abnormity, and calculating a distance fluctuation variance through a sliding window mechanism to perform secondary verification. And finally, processing and positioning anomalies by using image morphology, and dividing the degree of anomalies according to the relative deviation between the energy density and the mean value of the historical model. According to the method, the mahalanobis distance and the multivariate Gaussian distribution are introduced, secondary verification and abnormal positioning are combined, the limitation of a traditional method is overcome, the detection accuracy and reliability are effectively improved, the misjudgment and missing judgment rate is reduced, and the method does not depend on a large amount of labeled data and is high in practicability.

The invention discloses a multi-stage spatial-temporal clustering method and system based on a fused mahalanobis distance. The method comprises the following steps: acquiring a spatio-temporal data set, and determining a spatio-temporal neighbor relation of samples in the data set; calculating a space communication distance and a time decay distance of the sample; fusing the space communication distance and the time decay distance by using a mahalanobis distance to obtain a relative distance of the sample; selecting a class cluster center from the data set according to the local density of the sample and the relative distance; adopting a multi-stage distribution strategy to distribute non-class-cluster center samples to corresponding class clusters; wherein the multi-stage allocation strategy comprises an inevitable allocation stage based on space-time shared neighbor and a similarity allocation stage based on a weighted similarity matrix. According to the method, the key problems that an existing space-time clustering algorithm is insufficient in space-time attribute coupling processing and sensitive to distribution errors are solved, and the clustering accuracy, robustness and practicability in the fields of intelligent traffic analysis, seismic sequence recognition and the like are remarkably improved.

The invention relates to the field of network security, and discloses a problem line determination method based on IP address classification and identification, which comprises the following steps of: collecting public network IP behavior data in a network flow log, extracting multi-dimensional characteristics such as a time regularity entropy value, a protocol diversity proportion, target port dispersity and a request rate change rate, and determining a problem line according to the extracted multi-dimensional characteristics; constructing an IP association graph fusing feature similarity, time synchronism and physical position constraints; an abnormal IP group is identified by adopting a community division algorithm, and the detection sensitivity is adaptively optimized according to a real-time network load and a historical false alarm rate in combination with a dynamic threshold adjustment mechanism based on a PID feedback control model; the attack type is judged through protocol-port mapping and Mahalanobis distance statistics, alarm information containing an abnormal IP list, a time window and a classification result is generated, and the alarm information is linked with safety equipment to execute a defense strategy. According to the method, the defects of a traditional method in the aspects of multi-dimensional attack recognition, dynamic environment adaptation and cooperative attack detection are overcome.

The invention discloses an equipment abnormity monitoring method and system based on the Internet of Things, and relates to the technical field of intelligent operation and maintenance of the Internet of Things, and the method comprises the steps: collecting monitoring data to generate a high-dimensional original matrix, carrying out the optimization through employing a GCN model and combining with ACO, carrying out the updating through a comparison learning model and an FCM algorithm, and carrying out the searching of global optimum through employing a VAE model and combining with a PSO algorithm. The method comprises the steps of performing classification optimization based on K-means clustering and BSO, performing MLE calculation, updating dynamic causal KG through Granger causal test, generating a multi-modal result array through NSM, a scoring formula, a naive Bayesian model, a Mahalanobis distance formula and a logistic regression model, and performing optimization by using a fuzzy rule and GWO. According to the method, the GCN model is combined with the adaptive optimization algorithm, the precision and response speed of anomaly monitoring are improved, optimization is carried out by using the fuzzy rule and introducing the GWO based on multi-modal causal reasoning, and the reliability and efficiency of anomaly monitoring are improved.

The invention discloses a bearing degradation trend prediction method and system based on multi-domain feature dynamic fusion and dimensionality reduction, and the method comprises the steps: collecting full-life vibration signals of a bearing, synchronously marking three stages of health, degradation and fault, constructing multi-dimensional features such as a time domain, a multi-scale frequency domain, a time-frequency domain, and the like; evaluating the cross-stage difference of the features by using double criteria of mahalanobis distance and information entropy, and adaptively adjusting the weight to complete optimization; threshold cutting, linear proportion, Softmax or hierarchical weighting strategy empowerment are automatically selected according to data distribution, and energy is reserved through PCA for dimension reduction. And a TCN-GRU deep network model is constructed. Real-time data are input into the model to predict the degradation state, if errors exceed the limit, feature reconstruction and model retraining are triggered, and full-life-cycle high-precision high-robustness multi-stage continuous online monitoring is achieved. The method aims at solving the problems that the diagnosis precision is limited and the working condition adaptability is insufficient due to the fact that single time domain or frequency domain features are excessively depended and the features of each stage of fault evolution are difficult to comprehensively characterize.

The invention discloses a battery fault identification method based on probability labels and identification feature learning. The method is suitable for modeling and discrimination of various fault states in small sample and weak label scenes. The method comprises the following steps: firstly, acquiring key parameters such as voltage, current and temperature in an operation process of a battery system, and constructing standardized time sequence characteristic data; secondly, three types of pseudo labels are generated based on multi-source information such as alarm time difference, prediction residual error and mahalanobis distance, and a unified abnormal probability label is obtained through weighted fusion; constructing positive and negative sample pairs according to the difference between the tags, and introducing difficult samples with similar features but large tag difference to enhance the discrimination ability of the model; then constructing a twin neural network structure composed of shared parameter sub-networks, inputting positive and negative sample pairs for comparative learning, and extracting low-dimensional embedding features with clustering and distinguishability; and finally, through calculating a space distance between a new sample embedding vector and a known fault type, identification of a current fault type and evaluation of an abnormal degree are realized.

The invention provides a dual-antenna attitude and orientation and robust self-adaption method based on integrated navigation, which comprises the following steps: combining an INS (inertial navigation system) and a dual-antenna GNSS (global navigation satellite system), and enabling the system to output continuous and stable high-precision carrier attitude information when a GNSS signal is interfered by using dual-antenna integrated navigation. According to the method, most of satellite end clock correction, ionosphere and troposphere errors and receiver end clock correction are eliminated by using double-antenna GNSS pseudo-range and carrier phase double difference, robust statistics are calculated through heading prediction residual vectors to obtain robust factors, an observation noise covariance matrix is expanded to reduce the influence of system noise and observation noise, and the system performance is improved. A Mahalanobis distance based on a prediction residual vector is introduced to detect whether a system is abnormal or not, a state prediction covariance matrix is adjusted through a self-adaptive factor, weight reduction of an abnormal INS dynamic model is achieved, and the stability and reliability of system attitude information are improved through a noise covariance self-adaptive control mechanism.

The invention provides a long text intelligent review and prediction method fusing a dynamic knowledge evolution mechanism, and relates to the technical field of text review, and the method comprises the steps: carrying out the structural analysis of a long text, constructing an initial knowledge graph, and generating an evolution knowledge graph through combining a time sequence change mode of an entity relationship in a historical text; calculating semantic similarity between word vectors and graph embedding to realize information interaction; deep semantic features are extracted to calculate the mahalanobis distance between the deep semantic features and an abnormal category prototype to determine an abnormal mode; and combining historical evolution trajectory modeling time sequence characterization to predict an abnormal development trend. And the accuracy and prediction capability of long text review can be effectively improved.

The invention discloses an aero-engine group health evaluation method based on multi-working-condition dynamic clustering, and belongs to the field of aero-engine health state evaluation. The method comprises the following steps: firstly, carrying out clustering analysis on set parameters in engine operation data, and carrying out merging processing on small-scale abnormal clusters to obtain a working condition category division result; secondly, constructing a health baseline data set, carrying out standardized preprocessing on sample data in a working condition category division result, and carrying out nonlinear dimensionality reduction to obtain a low-dimensional feature data set; thirdly, performing clustering analysis on the low-dimensional feature data set by adopting a Gaussian mixture model, and calculating an average mahalanobis distance between a sample of each clustering category and a health reference center to obtain multi-level health levels corresponding to different clustering categories; and finally, through fusing the membership soft probability and the sample individual mahalanobis distance, constructing a continuous health score and obtaining a health grade determination interval. According to the method, health state characteristics under different working conditions can be effectively identified, individual difference modeling and group transverse comparison evaluation are supported, and the accuracy is improved.

The invention relates to the technical field of fault detection, in particular to an industrial fault detection method and system based on dynamic drift perception and diffusion enhancement. The method comprises the following steps: constructing an unsupervised fault detection DDA-DE model, and processing an industrial data flow by utilizing dynamic drift awareness DDA to establish a statistical distribution baseline and a drift threshold; determining an initial model parameter and an anomaly threshold by using diffusion enhanced anomaly detection DE; robustness enhancement of concept drift is carried out on the unsupervised fault detection model based on a diffusion strategy; according to the Mahalanobis distance real-time drift sensing algorithm based on industrial enhancement, the concept drift phenomenon can be detected more efficiently, and collaborative detection of data drift and abnormal events is achieved through the parallel design of the drift sensing algorithm and the fault detection classifier.

The invention discloses a CAN bus intrusion detection method based on adaptive unscented Kalman filtering, and the method comprises the steps: obtaining real-time message data, and carrying out the preprocessing of the real-time message data, and obtaining a time sequence feature vector; constructing a nonlinear state space model based on the time sequence feature vector; performing unscented Kalman filtering state prediction based on the nonlinear state space model; inputting the time sequence feature vector as an actual observation value, calculating a Kalman gain to correct an unscented Kalman filtering state prediction result, and outputting a state estimation residual error; dynamically updating a process noise covariance matrix through exponentially weighted moving average based on the state estimation residual, and adjusting a measurement noise covariance matrix according to the measurement innovation sequence; calculating the mahalanobis distance of the state estimation residual error, comparing the mahalanobis distance with a self-adaptive anomaly detection threshold value, and judging whether an intrusion behavior occurs or not; and if the abnormal score exceeds a threshold value, triggering a multi-level alarm mechanism, recording a suspicious message and executing a safety protection operation.

The invention provides a hydropower station AI supervision system and method based on a multi-modal large model, and relates to the technical field of intelligent hydropower. The system comprises a multi-modal data acquisition module, a cross-modal space-time alignment module, a multi-modal feature extraction module, a multi-modal large model processing module and an intelligent reasoning and decision module. A neural differential equation model is introduced to carry out space-time alignment on asynchronous sensing data, networks such as Vision Transformer, MelCNN, TCN and the like are utilized to extract multi-modal features, cross-modal fusion analysis is realized by combining a local and global attention mechanism and dynamic weight distribution, and equipment abnormality is further reasoned based on a reconstruction error, a mahalanobis distance and a knowledge graph and a maintenance strategy is generated. According to the method, high-precision anomaly detection, fault root cause positioning and dynamic maintenance optimization of key equipment of the hydropower station are realized, diagnosis errors caused by traditional manual inspection and data splitting are avoided, and the operation and maintenance intelligence level and the equipment operation reliability are improved.

The invention provides a self-adaptive robust vehicle navigation method and device based on LSTM assistance, and the method comprises the steps: calculating a process noise scale factor through process noise covariance estimation, and dynamically adjusting a process noise covariance matrix Q in a Kalman filter; executing a prediction process of Kalman filtering to obtain a one-step prediction state value and a state prediction covariance matrix; calculating a robust factor and a mahalanobis distance adaptive factor based on the prediction residual vector, and performing expansion adjustment on the prior measurement noise covariance matrix R and the predicted state prediction covariance matrix P; executing an updating process of Kalman filtering to obtain a state optimal estimation and a covariance matrix at the current moment; and when the GNSS loses lock, pseudo GNSS extended Kalman filtering is carried out by using a pseudo GNSS measurement value generated by the LSTM neural network and combining a process noise covariance estimation method. The method can better adapt to the dynamic characteristics of the navigation system in different environments, and provides powerful support for the stability and reliability of the combined navigation system.

The invention discloses an enterprise big data security early warning method based on anomaly detection, and the method comprises the following steps: S1, collecting original data, and carrying out the format unification and structure standardization processing; s2, preprocessing is carried out, and feature vectors are constructed; s3, a behavior entity relation graph is constructed, a graph attention network is adopted for training, and structural features in a normal behavior mode are learned; s4, calculating the deviation degree between the current behavior and the normal behavior; s5, reconstructing the feature vector, measuring the deviation degree between the current behavior and the standard behavior distribution by using a mahalanobis distance, and calculating the posterior anomaly probability of the behavior through a Bayesian updating mechanism; and S6, evaluating the risk level of the current behavior according to the posterior anomaly probability, and generating a corresponding early warning event. According to the method, the graph attention network and the Bayesian self-coding technology are fused, enterprise behavior anomaly detection and grading early warning are achieved, and the method has the advantages of being high in recognition precision, high in self-adaption and timely in response.

The invention discloses a network fault diagnosis method and system based on a 5G communication gateway, and relates to the technical field of network fault diagnosis, and the method comprises the steps: collecting the multimodal data of the 5G communication gateway, calculating the data features, and carrying out the abnormal data screening; the method comprises the following steps: performing data preprocessing to form a correlation matrix, extracting an observation vector of an event window, calculating a mahalanobis distance and making an abnormal mark to construct an initial weight vector, determining feature weighting output by using a steepest descent method, optimizing a feature weight, obtaining a fusion feature vector, selecting high-correlation features to determine an edge weight, and constructing a snapshot sequence of different time windows. According to the method, the observation vector in the event window is extracted, the mahalanobis distance is calculated, collaborative anomalies between the features can be captured from multiple angles, the feature weight is optimized through the steepest descent method, the model can pay more attention to the most important feature of anomaly detection when the mahalanobis distance is calculated, and the detection accuracy is improved. Therefore, the overall detection effect is improved.

A method for simultaneous localization and mapping underwater. A robot is equipped with an IMU inertial unit and a sonar unit. When the robot submerges underwater, a buoy connected to the robot floats on the water surface, and the buoy moves in coordination with the movement of the robot. The position and observed velocity of the buoy are obtained by a shore-based lidar. During motion estimation in a SLAM algorithm, when an angular velocity is below a preset angular velocity threshold, the observed velocity of the buoy is decomposed into x-axis velocity and y-axis velocity, and updated as the two-dimensional operating velocity of the robot. The polar coordinates of an obstacle in a map under the coordinate system of the robot are associated with the polar coordinates of sonar data of a current frame by using the Mahalanobis distance. The polar coordinates of a new obstacle are converted to the world coordinate system and added to the map. The present invention has the advantages of not relying on underwater visibility, not requiring pre-installed devices, and possessing good generalization and stability.

The invention relates to the technical field of test bed operation monitoring, in particular to a fuel pump test bed operation monitoring method and system. The method comprises the following steps: acquiring monitoring time sequence data of operation of a fuel pump test bed, performing feature extraction, constructing a coupling prediction model of a space-time diagram attention network-unscented Kalman filter, and predicting a process noise covariance matrix and a measurement noise covariance matrix according to the coupling prediction model, and performing state estimation by using an unscented Kalman filter to obtain vector posterior probability distribution, constructing a fault evolution trajectory manifold, calculating a mahalanobis distance between the vector posterior probability distribution and the fault evolution trajectory manifold, and taking the mahalanobis distance as a monitoring index. According to the scheme of the invention, the deep features which can better reflect the inherent nonlinear and complex dynamic characteristics of the system can be extracted from the multi-source data, the estimation accuracy of the filter on the potential health state of the system is improved, and the defects that model parameters are fixed and gradual change faults are difficult to capture in a traditional method are overcome.

The invention discloses a traditional Chinese medicine pesticide residue detection method and system based on artificial intelligence and a medium, particularly relates to the technical field of image processing and intelligent detection, and is used for solving the problem that the capability of distinguishing complex surface textures and pesticide residue areas of traditional Chinese medicines is insufficient. The method comprises the following steps: extracting local gradient direction distribution features of a high-resolution image, and combining gradient magnitude clustering processing to generate a gradient feature map of a texture edge contour; carrying out multi-scale fusion on the gradient features and the color channel data, and constructing a multi-channel feature map representing textures, colors and spatial distribution; analyzing quantitative feature relevance based on covariance of a high-order color moment and a directional entropy, and dividing normal textures and abnormal residual regions through a mahalanobis distance classifier; and performing spectral band matching degree evaluation on the abnormal region, fusing geometric morphological characteristics and spectral scores, inputting the fused geometric morphological characteristics and spectral scores into a pre-trained classification model, and generating a pesticide residue detection result by using a nonlinear correlation decision, thereby realizing accurate distinguishing between the natural texture on the surface of the traditional Chinese medicinal material and the pesticide residue.

The invention discloses an industrial robot predictive maintenance method and system based on multi-source data fusion, and the method comprises the steps: synchronously collecting vibration, current, temperature, acoustic and visual signals through multiple types of sensors, carrying out the filtering, correction and normalization processing, and constructing a multi-modal feature set; cross-modal alignment is realized through time compensation, after dimensionality reduction, a mechanical vibration group, an electrical performance group, a thermodynamic group and a motion precision group are divided, mahalanobis distances of the groups are calculated based on a historical health reference to serve as local anomaly degree scores, weights are dynamically adjusted according to the change rate, and the weights are combined into a preliminary health index. And introducing a nonlinear amplification mechanism to enhance high-value response, adaptively switching smooth intensity according to a degradation trend, and outputting a comprehensive health index. According to the method, comprehensive perception and dynamic evaluation of the operation state of the practical training platform are realized, multi-source heterogeneous information is effectively fused, the limitation of single signal monitoring is overcome, and the anomaly recognition accuracy is remarkably improved.

The invention discloses a real-time dynamic trajectory tracking method and system for millimeter wave radar gesture recognition, and relates to the technical field of gesture recognition tracking, and the method comprises the steps: receiving an echo signal reflected by a gesture, extracting a potential target point cloud, and carrying out the clustering generation of a gesture point cloud sequence; establishing a multi-modal motion model library, dynamically selecting an optimal motion model by adopting graph matching, and generating a prediction state in combination with a gesture point cloud sequence; on the basis of a Poisson multi-Bernoulli hybrid filtering framework, according to the signal-to-noise ratio and spatial distribution of the current gesture point cloud sequence, dynamically adjusting the observation weight, optimizing the observation point cloud, and carrying out optimal association by combining Mahalanobis distance with dynamic time warping; a multi-hypothesis tracking strategy is adopted to maintain trajectory hypothesis, an optimal trajectory is selected through a trajectory scoring mechanism, and Kalman filtering smoothing processing is performed on the optimal trajectory. According to the method, high-precision and low-delay tracking of gesture motion is realized, gesture habits of different users and complex environment interference can be adapted, and meanwhile, relatively high track precision is kept.

The invention relates to the technical field of power system monitoring and fault diagnosis, and particularly discloses a fault diagnosis system of an electrical variable measurement insulator detection device. The system comprises a synchronous acquisition module, a multi-physical field feature extraction module, a self-adaptive fault diagnosis engine and a feedback module. By synchronously acquiring voltage harmonic waves, leakage current and temperature data, extracting feature vectors of coupling electric-thermal influence and performing two-stage collaborative diagnosis by using a dynamic threshold value and a mahalanobis distance, high-sensitivity and self-adaptive accurate identification and early warning of an early latent fault of the insulator are realized. According to the system, by constructing a synchronous data acquisition module and a multi-physical field feature extraction module, two key influence factors, namely an operation voltage harmonic component and an environment temperature, are brought into a diagnosis system in a quantifiable feature form for the first time.

The invention discloses a smart building energy consumption data monitoring management method and system, and relates to the technical field of smart building and energy management, and the method comprises the steps: constructing a dynamic graph structure, extracting joint features through employing a graph attention mechanism, weighting the Mahalanobis distance of a node through employing a batch average attention coefficient, and obtaining an abnormal scene feature vector; a target classification function is defined, an EPC-PSO algorithm is used for optimization, and a Softmax classifier is used for classifying abnormal scene feature vectors; defining an energy consumption efficiency objective function, decomposing into a sub-problem of each device by using Lagrange, outputting a global initial strategy vector by using a gradient descent method, defining a smart building task, constructing a matrix of a comprehensive benefit weight, converting the device and the task into a bipartite graph problem, and solving by using a Hungary KM algorithm; the batch average attention coefficient weights the mahalanobis distance, the robustness of anomaly detection is enhanced, and the comprehensive benefit of resources is improved by using Lagrange decomposition, a gradient descent method and a Hungary KM algorithm.

The invention discloses a self-learning multi-target tracking method and system based on cross-modal perception, and the method comprises the steps: S1, collecting visible light and infrared video streams, carrying out the time sequence alignment, inputting a cross-modal fusion network based on Transform, deeply fusing the information of two modals through a cross attention mechanism, and generating a fusion feature map; s2, positioning a target by using a key-point-based anchor-frame-free detector Center Net, and extracting an identity re-identification Re-ID feature at the central point of the target; s3, adopting a parallel association and prediction process and a state adaptive predictor SAP module to perform motion state prediction on an existing track, and a confidence sequence associator CSA module to dynamically generate a decision confidence interval through online learning of statistical distribution of mahalanobis distances to perform decision judgment; and S4, introducing a global trajectory corrector GCM module, performing post-processing on trajectory interruption, and realizing trajectory stitching and identity ID correction. According to the invention, real-time tracking of multiple targets in a complex environment is realized.

The invention discloses an intelligent sand excavation supervision system based on multi-source data fusion, and relates to the technical field of machine learning, and the system collects target river reach data in real time through a multi-source sensing network module; the spatial-temporal feature fusion module generates a dynamic state fingerprint matrix; the adaptive baseline monitoring module establishes and dynamically updates a normal state baseline under multiple conditions, and triggers an abnormal disturbance alarm by calculating a mahalanobis distance between a real-time fingerprint and the baseline and combining collaborative deviation verification of acoustics, turbidity and water flow characteristics, and the multi-task analysis module adopts a parallel neural network architecture, so that a multi-task analysis result is obtained. The illegal operation type probability, the strength estimation value and the environment disturbance level are synchronously output; the three-dimensional visualization early warning module generates an early warning interface based on the analysis result; the dynamic knowledge management module and the self-adaptive optimization module are used for improving the analysis accuracy and continuously optimizing the system performance by using historical experience; the method has the advantages that abnormal disturbance events such as illegal sand excavation and the like can be accurately and intelligently supervised in real time, and powerful capability is provided.

The invention discloses an efficient storage and transmission method for monitoring data of a low-voltage power distribution network based on lossy / lossless hybrid compression, and relates to the technical field of data storage and transmission, comprising the following steps: completing data denoising correction at an edge node, and setting a plurality of compression strategies and layering mechanisms; judging whether the data is abnormal or not based on the mahalanobis distance, and selecting a proper compression mode through reconstruction error and bandwidth adaptation; the compressed data is subjected to importance labeling and FEC optimization and then sent to a receiving end, the receiving end evaluates the decoding quality and the packet loss rate, and finally a feedback result is used for online updating of the auto-encoder. According to the method, the differential FEC redundancy rate is allocated, so that the lossless fidelity of a key fault waveform and the high compression ratio of a common periodic signal are considered; and meanwhile, closed-loop self-adaption of compression discrimination, coding and model optimization is realized by utilizing online updating of the variable auto-encoder, so that the storage and transmission efficiency is improved, and the reliability and the real-time performance of the system in sudden failure and network fluctuation scenes are enhanced.

The invention discloses a flight training quality evaluation method based on DTW and K-means clustering, and belongs to the technical field of pilot training, and the method comprises the following steps: S1, data collection and processing; s2, data preprocessing and standard template generation; s3, DTW similarity measurement and mahalanobis distance weighting are carried out; s4, performing similarity measurement and clustering analysis; s5, generating a personalized training scheme; therefore, the invention provides a scientific, objective and efficient flight trainee quality evaluation method, the landing process of the pilot is subjected to comprehensive quantitative analysis in a data-driven manner, a more accurate and personalized flight training scheme is provided, the flight level of the flight trainee is finally improved, and the risk of flight accidents is reduced.