327 results about "Spectral clustering" patented technology

The invention provides a virtual DPU power plant simulation fault restoration method and system based on digital twinning, and relates to the technical field of digital twinning, and the method comprises the steps: carrying out the preprocessing of collected DPU power plant operation data, including noise reduction, time sequence alignment and abnormal point elimination, carrying out the data quality evaluation, training a fault feature mapping model based on the processed data, and carrying out the fault restoration of the DPU power plant. The model is used for recognizing abnormal clusters in real time, a fault evolution path is searched and determined in combination with a conditional random field and a Monte Carlo tree, an optimal path is determined through a particle filtering algorithm, fault root causes are determined in combination with causal analysis, spectral clustering and a Bayesian network, and a fault diagnosis report is generated.

The invention provides an urban functional area gridding identification and classification method based on multi-source spatio-temporal data, and relates to the technical field of area identification, and the method comprises the steps: carrying out the dynamic grid division of an urban area through employing a self-adaptive grid division algorithm, obtaining the multi-source spatio-temporal data of a gridding area, carrying out the space-time alignment, and generating a multi-dimensional time sequence feature set; extracting multi-dimensional feature vectors of the grid units through a multi-modal feature fusion algorithm, acquiring long-range time sequence features by using a bidirectional gating recursion unit, calculating spatial correlation features based on a multi-head attention mechanism, and inputting the two types of features into a feature fusion network with jump connection to generate enhanced spatial-temporal features; and an optimization strategy is adopted, a functional area clustering center point is determined through density peak value clustering in sequence to obtain an initial division result, the initial result is optimized by combining spectral clustering of spatial adjacency constraints, and the boundary of the functional area is optimized by using a simulated annealing algorithm, so that accurate identification and classification of the urban functional area are realized.

The invention relates to the technical field of data protection, and discloses a privacy protection type data joint modeling method based on federated learning, which comprises the following steps: acquiring local data to perform meta-feature extraction, calculating key statistics to characterize data characteristics, collecting meta-features, grouping the meta-features into similar feature clusters through spectral clusters, and carrying out feature clustering on the similar feature clusters; dynamically allocating and calculating resource weights according to the similar characteristic cluster scale and the equipment computing power; distributing a basic privacy budget according to the client type, calculating a local model accuracy rate and an intra-cluster level difference, dynamically adjusting the privacy budget, adding adaptive Gaussian noise based on the privacy budget, and adjusting gradient sensitivity of gradient calculation; verifying gradient compliance through zero knowledge, carrying out safe aggregation on gradients passing verification, optimizing a meta-model through a knowledge distillation loss function, and generating confrontation sample analysis to obtain a leakage risk value to identify knowledge leakage risks; sensitive neurons in the neuron sensitivity positioning element model are analyzed and calculated, directional noise is injected, and initial parameters are adjusted for initialization training.

The invention discloses a power grid dispatching anti-error method and system based on topology analysis, and relates to the technical field of power grid dispatching anti-error, and the method comprises the following steps: building a remote signaling credibility evaluation model based on remote signaling evaluation data, and obtaining high-credibility remote signaling data; performing sequential logic verification based on high-credibility remote signaling data and multi-source data fusion to generate a power grid topology model; carrying out partition division on the power grid topology model based on spectral clustering and a state-driven edge weight method, and carrying out boundary equivalence processing to obtain a plurality of topology modules; and performing simulation scheduling operation on the plurality of topology modules based on a scheduling instruction dispatching order objectification analysis result, performing path traversal based on a depth-first search method, and performing anti-error analysis on the traversed path in combination with a reinforcement learning method. According to the power grid dispatching anti-error method combining remote signaling credibility evaluation, dynamic topology partition and reinforcement learning algorithm, the anti-error verification capability and the intelligent level of a power grid dispatching system are comprehensively improved.

The invention discloses a voiceprint clustering method guided by a speaker recognition pre-training model based on short voices, and the method comprises the following steps: obtaining an original voice signal, and randomly combining a plurality of enhancement strategies to achieve data enhancement; performing voice segmentation on the voice signal after data enhancement based on a uniform segmentation mode; extracting voiceprint features of the segmented voice based on an attention mechanism of global time-frequency domain context modeling; and obtaining a clustering result based on K-means clustering and spectral clustering, matching the clustering result with a real speaker tag, performing reverse transmission based on angle-dependent AAM-Softmax loss, and outputting a voiceprint clustering result. According to the method, the influence of environmental interference on feature extraction can be overcome, feature dimensions which are more effective for identity identification can be screened, a clustering output effect which is superior to that of a mainstream algorithm can be obtained with a relatively low parameter quantity, and the robustness under a noise interference condition is improved.

The invention discloses an electricity larceny identification method based on graph calculation, and particularly relates to the technical field of electricity utilization anomaly detection of an electric power system. Historical power consumption data and a power supply topological relation of power consumers are collected, and a multi-dimensional behavior graph model fusing behavior characteristics and structural information is constructed; performing structure disturbance analysis on each node in the graph, calculating information entropy change before and after node removal, performing attention fusion on a time sequence behavior feature of the node and a structure disturbance vector, constructing a joint feature vector, and enhancing feature expression through spectral clustering and linear reconstruction; a behavior propagation field and a disturbance adjustment mechanism are introduced into the graph to form a disturbance response graph, and an abnormal gathering area is identified through path energy analysis and focusing area fitting; calculating confidence scores of the nodes and outputting a suspicious user list; the method can realize efficient identification of electricity stealing behaviors with strong concealment and complex transmissibility, and has the advantages of high precision, strong interpretability and wide application scene adaptability.

The invention discloses a networking charging data cross-platform fusion and transmission method based on an intelligent scheduling algorithm. The method comprises the following steps: S1, collecting and preprocessing networking platform charging data; s2, feature similarity is calculated, and an undirected graph with samples as nodes and similarity as edge weight is constructed; s3, calculating a standardized Laplacian matrix, extracting feature vectors, and clustering to generate feature cluster tags; s4, constructing a modal mapping matrix, mapping the feature vectors, and inputting the mapped feature vectors into a first convolution restricted Boltzmann machine to extract fusion features; s5, inputting the fusion features into a second convolution restricted Boltzmann machine, and combining graph structure sorting and constructing a weighted directed graph; s6, initializing a virtual scheduling factor in the graph, and searching an optimal transmission path by an ant colony algorithm; and S7, distributing fusion features according to paths, and realizing cross-platform efficient transmission of charging data. According to the method, spectral clustering, modal mapping, convolution extraction and intelligent scheduling are fused, and the multi-platform charging data fusion accuracy and transmission efficiency are improved.

The invention discloses a high-precision distributed photovoltaic output prediction method and system based on micrometeorology and feature hierarchical clustering, and relates to the technical field of distributed photovoltaic output prediction.The method comprises the steps that numerical weather forecast data are collected, an initial micrometeorological field is generated through space-time alignment and self-adaptive KNN interpolation, and the initial micrometeorological field is subjected to feature clustering; a WRF-LES system and a bidirectional LSTM are combined to establish cross-scale mapping, a dynamic residual correction field is fused to generate hectometer-level high-resolution micrometeorological data, and the problem of insufficient resolution of traditional numerical forecasting is solved. MIC and PA-DTW are used for jointly analyzing the characteristics of the power station, and dynamic clustering is achieved through a sliding time window and incremental spectral clustering. According to the method, a physical information graph network and causal expansion convolution are coupled to extract features, federal learning cross-power-station cooperative training is combined, the distributed photovoltaic output prediction precision and robustness are improved, and privacy security is considered.

The invention provides an artificial intelligence advertisement pushing method based on multi-dimensional historical data analysis, and the method comprises the steps: constructing second-level, minute-level, hour-level and other behavior feature views of different time granularities for extracted original features, and forming multi-time scale feature representation; analyzing feature views under different time scales through a graph mining algorithm, mining an association rule between feature nodes, and constructing a cross-scale feature association graph; in the feature correlation map, according to the spectral clustering result of the feature nodes and the gravitation direction of the feature edges, the temporal correlation degree of different feature combinations is judged; if the time correlation degree of the feature combination exceeds an empirical threshold value, endowing the time weight of the feature combination with a high weight coefficient of power law distribution; and dynamically adjusting the network depth and the connection density of the association atlas according to the vertical dimension and the semantic hierarchy of the features, and optimizing the feature weight distribution in the atlas.

The invention discloses a high and cold arid region slope soil stability safety risk evaluation system and method, and relates to the technical field of slope engineering. The state of each side slope under multi-dimensional indexes such as freeze-thaw cycle frequency, dry-wet alternation index, wind erosion strength, shear strength, water content, porosity and fracture density is quantified, the feature similarity between any two side slopes is calculated, and then a side slope feature similarity matrix is formed. Based on the matrix, an unsupervised clustering algorithm (such as spectral clustering, similarity propagation and the like) can be adopted to divide a plurality of side slopes into similar subsets with structural characteristics similar to environmental response, and category attribution of risks is achieved. On the basis, structural variation analysis, historical instability statistics and central risk difference extraction are performed on similar slope samples, so that the internal instability tendency of the slope can be identified, and a risk prediction model suitable for the type of slope can be constructed through feature training.

The invention relates to the technical field of cloud platform computer control, and discloses a computer control system and method based on a cloud platform. The system comprises a multi-source data acquisition module which acquires heterogeneous data in real time and divides the heterogeneous data into time sequence segments; the data fusion preprocessing module is used for removing noise and screening key features to generate a standardized data stream; the dynamic decision control module outputs a control instruction set based on a deep reinforcement learning algorithm; the abnormal mode recognition module is used for recognizing anomalies by adopting spectral clustering and an isolated forest model; the resource flexible scheduling module is used for optimizing resource scheduling by using an improved genetic algorithm; and a safety audit module is also arranged to guarantee instruction safety. According to the system and the method, heterogeneous data of the cloud platform can be effectively acquired and processed, accurate decision control is realized, abnormity is accurately identified, resources are reasonably scheduled, system security, stability and resource utilization rate are improved, and complex service requirements of the cloud platform are met.

The invention relates to the technical field of tunnel defect identification, in particular to a tunnel defect identification method based on a dielectric distribution diagram. According to the method, on the basis of AI rock-soil perspective radar field monitoring, gprMax simulation and laboratory electromagnetic data, a two-dimensional dielectric distribution diagram is generated through variational Bayesian inversion fusion. Through multi-scale spectral clustering and expert knowledge, a potential abnormal region is automatically identified, and then accurate classification and identification of a defect region are realized by adopting graph form constraint propagation and integrating a Transform-graph neural network model. And finally, projecting an identification result to an original image, and generating a visual defect labeling layer and a structured report. According to the invention, high-precision, automatic, visual and structured detection of tunnel structure defects is realized, the intelligent level and risk early warning capability of tunnel safety operation and maintenance are significantly improved, and the digitization and intelligent process of tunnel operation and maintenance management is promoted.

The invention discloses a joint optimization clustering distributed photovoltaic power prediction method and system, and relates to the related field of distributed energy management, and the method comprises the steps: carrying out the local calling of data, carrying out the time sequence alignment, and constructing a multi-dimensional feature association number table; extracting geographic position information and historical output data to perform similarity analysis, performing spectral clustering, and outputting a photovoltaic cluster; collecting meteorological time series data according to the power prediction scale, inputting the meteorological time series data into the cluster power prediction model, and performing cluster power prediction; grid-connected space-time joint optimization is carried out according to the cluster-level power prediction value, and an energy storage charging and discharging scheduling strategy is output; after single-station fitting conversion is carried out and a distributed power prediction value is output, output deviation identification is carried out, and a fault node is positioned; and dynamically updating the energy storage charging and discharging scheduling strategy. The technical problems that existing power prediction is not high in precision, cross-cluster power optimization scheduling is difficult to achieve, and the power grid operation stability is poor are solved, and the technical effect of improving prediction accuracy and the power grid operation stability is achieved.

The invention provides a hybrid wireless eye movement tracking method and system based on multi-sensor fusion, electronic equipment and a storage medium, and relates to the technical field of eye movement tracking. The method comprises the following steps: firstly, acquiring various data by using various different types of sensors such as a Q-Var sensor, an inertial measurement unit and a high-speed panoramic camera; secondly, preprocessing the various sensor data by means of graph Fourier transform, a Kalman filter, a spectral clustering technology and the like; fusing the preprocessed sensor data based on a graph theory method; and finally, performing fixation track prediction based on a convolutional neural network and a gating circulation unit. According to the technical scheme, the performance of eye movement tracking under different environment conditions is improved, the eye movement tracking precision is higher, and the stability is better. In addition, the technology of the application also uses a 5G millimeter wave technology or an IEEE 802.11 ac protocol to transmit gaze tracking data, thereby achieving the purpose of bandwidth enhancement and low-delay data transmission.

The invention discloses a method for constructing an evaluation model for the influence of climate change on biodiversity, and particularly relates to the technical field of ecological modeling and climate response analysis. The method comprises the following steps: generating a high-resolution characteristic grid based on microclimate disturbance data and a remote sensing image, extracting ecological plaques and establishing a heterogeneity spatial index model, constructing a connectivity map and a coupling response path library, extracting a typical species response mode by using nonlinear dimension reduction and spectral clustering, and performing dynamic fitting in combination with biological survey data. And finally, generating a diversity attenuation trend prediction curve and a risk thermodynamic diagram, identifying a biodiversity collapse critical point and outputting an intervention priority, and the method can be widely applied to regional ecological early warning and protection planning.

The invention discloses a winter wheat LAI inversion method based on scenarized parameter calibration and ensemble learning. The method comprises the following steps: fusing a remote sensing image with auxiliary geographic information such as terrain and soil, and performing typical planting scene division on a research area by adopting a spectral clustering method; then scene optimization is carried out on highly sensitive parameters (such as LAI, Cab, Cm and ALA) in the radiation transfer model PROSAIL based on a Markov chain Monte Carlo (MCMC) algorithm, and parameter posterior distribution more consistent with actual observation is obtained; further utilizing optimization parameters to generate a high-quality and homogenized simulation sample set, and mapping the simulation sample set to a Sentinel-2 waveband to construct a training data set; and finally, performing training by adopting various machine learning models (such as LightGBM, XGBoost, RF and KNN), and constructing a fusion model through a stacking integration strategy to realize high-precision inversion of LAI. Experimental results show that the method has higher fitting precision and stronger generalization ability in a plurality of typical planting scenes.

The invention discloses a decision scheme generation method and device based on complex multi-modal management data, equipment and a medium, and relates to the technical field of data processing, and the method comprises the steps: receiving enterprise management data of texts, numerical values and images, extracting a unified feature vector, carrying out the clustering analysis through the dynamic updating of torque clustering and spectral clustering, and obtaining a clustering result; and a causal knowledge graph is constructed according to the result, and a decision scheme is generated by using a fourth-order inference engine, so that efficient processing and real-time response of an enterprise to dynamic multi-modal management data are realized, the cross-modal data analysis capability is enhanced, and the decision accuracy is improved.

The invention discloses a wireless communication resource allocation method based on dynamic spectrum sensing. The method comprises the following steps: acquiring equipment node data and network environment parameters, and performing preprocessing to generate feature data; based on the preprocessed feature data and the time window, generating a spatio-temporal feature map by adopting dynamic spatio-temporal feature coding so as to capture spatio-temporal association between the devices; a hyperedge set is constructed through a hyperedge dynamic generation mechanism by using equipment position information, and network topology changes are reflected; based on the hyperedge set, the transmitting power and the channel gain, a high-order interference tensor is constructed and decomposed, and the interference relation is accurately quantified; according to the interference tensor, the number of resource blocks and the service quality requirement, generating an optimal resource allocation strategy by adopting spectral clustering and an optimization algorithm; according to the method, through the space-time attention mechanism and the high-order interference tensor modeling, the dynamic adaptability and optimization performance of resource allocation can be improved, and an efficient solution is provided for a modern wireless communication network.

The invention discloses a wind power plant power dynamic optimization method and system based on spectral clustering partition, and the method comprises the steps: building a long-time scale wind power plant power maximization steady-state optimization problem based on an engineering wake flow model according to the average wind speed and wind direction, and solving an optimal yaw angle; calculating a wake flow coupling relation of the wind power plant in the current wind direction according to the optimal yaw angle to construct a wake flow diagram; clustering the wake flow diagram based on a spectral clustering method, and dividing the wind power plant into a plurality of sub-regions; and constructing a short-time scale wind power plant power dynamic optimization problem with optimization targets including maximum wind power plant total power and minimum wind turbine generator thrust fluctuation for different subareas, and solving the short-time scale wind power plant power dynamic optimization problem to obtain the optimal control quantity of each generator. The wind power plant maximum power output optimization method aims at overcoming the defects of a traditional static optimization strategy in a dynamic environment, and the overall generating capacity of the wind power plant is dynamically improved.

The invention discloses a fast fuzzy spectral clustering method based on an anchor point diagram, and is applied to the technical field of fast spectral clustering. Comprising the following steps: acquiring an original data matrix, a category number and an anchor point number, generating anchor points, and constructing a kernel-free similar matrix and a kernel-free adjacent matrix; initializing an original data class guidance matrix and an anchor point data class guidance matrix; updating the class guidance matrix of the original data; updating the anchor point data class guidance matrix; and repeatedly updating until convergence, and outputting a fuzzy membership matrix. According to the method, the membership relationship between each original data point and different clustering centers can be identified more accurately, the interpretability of clustering is enhanced, the sensitivity of clustering to the original clustering centers is reduced, and the clustering precision is improved.

The invention relates to a multi-time scale scene analysis-based micro-grid day-ahead and intra-day optimal scheduling method, which belongs to the field of micro-grid scheduling, and is characterized in that a variational mode decomposition (VMD)-long short-term memory network (LSTM) multi-scale prediction framework is constructed, ultra-short-term precision is improved through variable mode decomposition and frequency division prediction, a Canopy-spectral clustering-K-means hybrid algorithm is designed, and the optimal scheduling of a micro-grid is realized. A typical scene is generated based on Latin hypercube sampling (LHS), the scene coverage capability is enhanced, a day-ahead and intra-day two-stage optimization model is finally constructed, a high-dimensional problem is rapidly solved by adopting a mixed integer programming algorithm, and theoretical support is provided for high-proportion renewable energy consumption and micro-grid refined scheduling.

The invention discloses an intelligent substation virtual loop online transmission verification method based on GOOSE / SV closed loop test, and relates to the technical field of intelligent substation operation and maintenance. Through precise clock synchronization and an improved cross-correlation algorithm, in combination with wavelet noise reduction and spectral clustering analysis, nanosecond synchronization quality evaluation of GOOSE / SV signals is realized, and the hidden transmission risk discovery time is shortened from regular maintenance to real-time monitoring; a hierarchical hidden Markov model is adopted to analyze equipment-level to system-level behavior modes, real-time probabilistic reasoning is realized in combination with a dynamic Bayesian network and particle filtering, a multi-dimensional evaluation system is constructed, the reliability evaluation capability under complex working conditions is remarkably improved, an optimization scheme is generated based on network path characteristic analysis and bottleneck identification, and the reliability of the system is improved. Multi-scene closed-loop verification is carried out by means of a digital twin technology, safety and reliability of parameter optimization are ensured, and full-process intelligent operation and maintenance of the virtual circuit of the intelligent substation from state perception to optimization verification are realized.

The invention discloses an intelligent customer service intention layered recognition method oriented to multi-modal interaction. The method comprises the following steps: acquiring and preprocessing three modal data of a text, a voice and an image; constructing a high-dimensional topological structure of modal data by using a topological data analysis method, and carrying out feature fusion; a causal relationship graph of modal features is constructed through a structural equation model, and causal contribution significant features are screened; a cross-modal self-supervision consistency regularization method is adopted, and an intention recognition model is trained under the condition of extremely few supervision data; and realizing multi-level layered classification output of the intention based on a spectral clustering method. The intention recognition accuracy and generalization performance of the intelligent customer service system are remarkably improved, and the method is suitable for complex multi-modal interaction scenes.

The invention relates to the technical field of air tightness detection, in particular to a diaphragm type energy accumulator air tightness detection method which is used for solving the problems that in the prior art, equipment operation characteristics cannot be accurately described, typical defect modes cannot be recognized in combination with spectral clustering, and a quantitative basis cannot be provided for equipment fault trend analysis and intelligent maintenance. The method comprises the following steps: constructing a low-dimensional state map to describe equipment operation characteristics, identifying typical defect modes in combination with spectral clustering, establishing a nonlinear correlation model to reveal a defect evolution relationship, optimizing classification model parameters by adopting an evolutionary algorithm, improving the identification accuracy, mining a most probable defect evolution path based on a transition probability matrix, and improving the identification efficiency. And a quantitative basis is provided for equipment fault trend analysis and intelligent maintenance.

The invention discloses a multi-view clustering method based on tensor feature extraction, and the method comprises the steps: inputting a multi-view data matrix, constructing a similarity matrix of each view through a K-NN algorithm and a Gaussian kernel function, and carrying out the spectral clustering to obtain a sample embedding matrix; performing singular value decomposition on an original data matrix of each view, taking first c left singular vectors to construct a feature embedding matrix, applying 2, 1 norm group sparse constraint on the feature embedding matrix, and connecting a sample embedding matrix through a bigraph to extract features; and normalizing the sample embedded matrix, and reconstructing a block diagonal matrix into a third-order tensor. And integrating the sample embedding matrix, the feature embedding matrix and global tensor learning to construct a target function, and optimizing through an alternating direction multiplier method until convergence. And finally, the normalized samples are embedded into the matrix to form block diagonals to form a consistent similarity graph, and a clustering result is obtained by using an N-Cut or k-means algorithm.

The invention discloses an active fluctuation collaborative stabilizing method and system for a high-proportion distributed new energy power grid, and relates to the technical field of power grid dispatching. According to the method, a physically consistent weather prediction model is established through multi-source meteorological data fusion, and a regional fluctuation propagation rule is accurately captured; identifying a high-risk fluctuation cluster based on dynamic time warping and spectral clustering, simulating a fluctuation propagation path by using a digital twin platform, and quantifying resource requirements; energy storage resource configuration is optimized by adopting mixed integer programming and a column generation algorithm, and multi-dimensional stability verification is carried out through a digital twin environment; a self-adaptive optimization mechanism based on reinforcement learning is established, continuous evolution of the system is realized, the technical bottlenecks of a traditional method in the aspects of fluctuation perception, resource allocation, system self-adaption and the like are solved, a collaborative stabilization mechanism with accurate prediction, intelligent recognition and decision optimization is formed, and a complete solution is provided for safe and stable operation of a high-proportion new energy power grid.

The invention discloses a historical building digital monitoring protection system based on multi-scale features. The system comprises a building data acquisition module, an original data optimization module, a feature extraction model construction module, a state recognition model construction module and a building digital monitoring module. The invention relates to the technical field of historical building digital monitoring, in particular to a historical building digital monitoring protection system based on multi-scale features, and the method comprises the steps: obtaining original data through building data; a data optimization method of data alignment, data cleaning, data standardization and data set segmentation is adopted; a deep learning model is adopted as a feature extraction model, and deep correlation features of building time-varying behaviors and structural damage are learned through feature decoupling driven by time sequence modeling and physical laws; an improved clustering model is adopted as a state recognition model, and robust recognition of the building state is achieved by introducing physical threshold constraint spectral clustering, a dynamic time kernel function and a multi-scale probability fusion mechanism.

The invention relates to the field of battery packs, in particular to a method and a system for judging a voltage difference fault of a battery pack. The method comprises the following steps: acquiring a voltage value of any battery module in a battery pack at a target moment, constructing graph structure data for the battery pack, and clustering the graph structure data by utilizing spectral clustering to obtain a plurality of clusters; obtaining an adjacent cluster, matching the adjacent cluster with the target moment cluster, and determining a value of a judgment function; calculating a pressure difference abnormal value of the battery module according to the calculated pressure difference variable quantity; and obtaining the accumulated value of the abnormal voltage difference value of any battery module in the preset sampling period, and taking the maximum value of the accumulated value in the battery pack as the abnormality degree of the battery pack to complete the fault judgment of the battery pack. According to the technical scheme of the invention, the precision of the voltage difference fault judgment result of the battery pack can be improved, and the operation safety of the battery pack is ensured.

The invention relates to the field of urban traffic management and intelligent traffic, and discloses a macroscopic region division method based on urban traffic demands, which comprises the following steps: acquiring traffic node traffic demands of an urban dynamic road network, and constructing a traffic demand adjacency matrix, a degree matrix and a Laplacian matrix, utilizing a spectral decomposition method to extract aggregation degree features among the nodes; a K-means clustering algorithm is adopted to classify the node feature vectors, and a macroscopic region is preliminarily divided; gPS normalized coordinate features are introduced, node space distribution is modeled by using two-dimensional Gaussian distribution, and a comprehensive distance optimization region division result is calculated in combination with a spectral clustering feature distance; and finally, iteratively optimizing category distribution based on the comprehensive distance to obtain a macroscopic region division scheme with the minimum segmentation loss. According to the method, the rationality of traffic area division can be improved, the macroscopic area coverage range is optimized, the scientificity of traffic management is enhanced, and the urban traffic resource allocation efficiency is improved.

The invention belongs to the technical field of image analysis, and particularly relates to a railway track image abnormal region segmentation system combined with fractal dimension feature extraction, which comprises a preprocessing unit, a multi-scale fractal feature processing unit, an abnormal region processing unit and a segmentation unit, the preprocessing unit is used for suppressing noise, retaining a main structure and outputting a denoising result; the multi-scale fractal feature processing unit is used for obtaining a fused fractal dimension based on a denoising result, and forming a feature vector by the fused fractal dimension and a denoising result gradient; the abnormal region processing unit is used for obtaining processed candidate abnormal regions by using a spectral clustering method; and the segmentation unit is used for performing intersection operation on the processed candidate abnormal region and a track mask to obtain a track abnormal region. According to the invention, high-precision segmentation of the abnormal region in the track image is realized.