163 results about "Density based clustering" patented technology

The invention discloses a dynamic road segment division based vehicle route guidance method, which is characterized in that a dynamic rod network connected graph is built through dynamic road segment division for the purpose of searching optimal path search and realizing dynamic navigation. The method specifically comprises the following steps: first, acquiring vehicle real-time information through a vehicle networking technology by the traffic center, and utilizing an algorithm of Density-based Spatial Clustering Of Applications With Noise (DBSCAN) to regularly and dynamically divide the regional rods, so as to generate the dynamic rod network connected graph; secondly, sending the position and destination of a vehicle itself to a traffic information center for asking for the optimal path; and finally, generating the optimal path on the dynamic rod network connected graph through utilizing a shortest path algorithm by the traffic information center according to the position and destination of the vehicle, and sending the information to the vehicle and realizing path guidance. The method has the advantages that the generated dynamic rod network connected graph which is accurate and real-time can provide the optimal path guidance for a traveler, thereby alleviating city traffic jam and improving running efficiency.

Embodiments are directed to a density-based clustering algorithm that decomposes and reformulates the DBSCAN algorithm to facilitate its performance on the Map-Reduce model. The DBSCAN algorithm is reformulated into connectivity problem using a density filter method and a partial connectivity detector. The density-based clustering algorithm uses message passing and edge adding to increase the speed of result merging, it also uses message mining techniques to further decrease the number of iterations to process the input graph. The algorithm is scalable, and can be accelerated by using more machines in a distributed computer network implementing the Map-Reduce program.

The invention discloses an indoor passive positioning method based on channel state information and a support vector machine. The method comprises the following steps: firstly preprocessing the acquired channel state information data, performing de-noising and smoothness through the adoption of a density-based spatial clustering of applications with noise and a weight-based moving average algorithm, and then using the principal component analysis algorithm to extract the features. The data after the preprocessing and feature-extracting can reflect the signal change more accurately and the dimension is greatly reduced. The passive positioning adopts two-stage positioning. In the training stage, the large positioning space is divided into sub-regions, the support vector machine classification and regression model is established for each sub-region so as to acquire a statistic model for accurately representing the nonlinear relationship between the position and the signal. The two-stage positioning firstly determines the sub-regions through the classification of the support vector machine, and the precision position is determined in the sub-region through the regression of the support vector machine. The method disclosed by the invention has the beneficial effects that the passive positioning can be performed in the absence of the active participation of the target, and the indoor positioning precision is improved to sub-meter level.

The invention discloses a method for identifying a flow characteristic curve of a governing valve for a steam turbine through a DCS (Distributed Control System) data mining technique. The method for identifying the flow characteristic curve of the governing valve for the steam turbine through the DCS data mining technique comprises the following steps of a, selecting sampling points relevant to the flow characteristic analysis of the governing valve, and obtaining DCS data; b, dividing the DCS data into data mining objects, auditing abnormal values in the data mining objects, and calculating the average value, the range and the slope of each sampling point of the data mining objects; c, carrying out dimensionality reduction on the range and the slope of each data mining object by applying a principal component analysis method; d, carrying out clustering on the running state of the governing valve by applying DBSCAN (Density-Based Spatial Clustering of Applications with Noise); e, correcting the data mining objects to rated boundary parameters by applying variable working condition characteristics of the steam turbine, a category-I correction curve and a category-II correction curve; f, fitting the data mining objects by applying a moving least-square method to obtain the flow characteristic curve of the governing valve for the steam turbine. The method for identifying the flow characteristic curve of the governing valve for the steam turbine through the DCS data mining technique can be used for identifying the flow characteristic curve of the governing vale of the steam turbine under the rated boundary parameters, and can be used for the parameter setting and optimization of a controller, so as to guarantee a generator set to run safely, stably and economically.

The invention provides an invasion detection method based on channel state information and a support vector machine. No special hardware facility is needed, an existing wireless network is fully used, and a common business router is used to realize security monitoring function. The coverage scope is wide, and privacy exposure can be prevented. The invasion detection method includes the steps of after obtaining CSI original data, conducting clustering and de noising for the subcarrier data in a channel by using a density-based clustering algorithm DBSCAN, smoothing the denoised data by using weight-based sliding average algorithm, and extracting characteristic values for data by using major constituent analyzing algorithm after data pre-processing. Data subjected to pretreatment and feature extraction can more accurately reflect the main change of signals and greatly reduce number of dimensions. The invasion detection precision is improved and the calculating complexity is reduced. The method uses an SVM classification algorithm to obtain a statistics model of non-linear dependence relation between an invasion state and a signal fingerprint, thereby achieving the purpose of invasion detection.

The invention relates to a method for realizing operating state analysis and fault diagnosis of a photovoltaic array based on a density-based clustering algorithm. The method comprises the following steps: firstly collecting a plurality of electrical parameters of a maximum power point of a photovoltaic power generation array during daily work so as to obtain an electrical parameter sample combination per day; normalizing the electrical parameter samples so as to obtain a test sample combination; calculating the normalized test sample combination so as to obtain a distance matrix; automatically clustering the test samples by adopting the density-based clustering algorithm so as to obtain a plurality of clusters; respectively calculating the minimum distance between each group of reference data and each cluster based on reference data obtained by a simulation model in advance so as to form a distance vector; and finally, comparing each element in the distance vector with a cutoff distance in the clustering algorithm, and identifying a work type to which each cluster belongs. According to the method disclosed by the invention, accurate fault diagnosis can be directly realized by clustering the daily operation data of the photovoltaic system.

The invention discloses a method for identifying and repairing power load abnormal data based on density clustering and LSTM, and belongs to the technical field of power quality analysis methods. According to the method, a density-based clustering algorithm (Density-based Spatial Clustering of Applications width Noise) and Long Short-Term Memory Neural Network are combined to identify and repair abnormal data. The method comprises the following steps: firstly, carrying out density clustering on data in units of days by utilizing a DSCAN algorithm to obtain abnormal data; then, using a long short-term memory (LSTM) neural network, taking the time series data judged to be abnormal as input of the LSTM neural network, and using the first n pieces of sequence data to predict the next piece ofsequence data; finally, the predicted value of the LSTM serving as an accurate value, setting an up-down floating threshold value is set, if the measured value exceeds the threshold value range, regarding the measured value as an abnormal value, and the predicted value of the LSTM serving as a correction value. According to the method, the time sequence and regularity of the power quality monitoring system data in the actual power grid are fully considered, the specific abnormal value can be accurately detected and repaired, and the method has good actual application value.

The invention discloses a dynamic object tracking method used for a port autonomous-driving vehicle. Firstly, continuous-frame point clouds are fused and merged through lidar installed at a truck headand GPS positioning information to obtain relatively dense point cloud perception data; then the fused point cloud data are voxelized to form a grid 3D graph, and randomness of a point cloud is reduced; then height data of voxel 3D information is used to eliminate the ground, a previous frame of fused data are used at the same time to obtain difference, most static obstacles are eliminated, and only dynamic object information is retained; then clustering is carried out on a remaining point cloud through a density-based clustering algorithm to obtain moving target information of a current frame; and finally, moving targets of adjacent frames are used to carry out a heuristic tracking algorithm to calculate a center, a size, speed, a movement direction, a life cycle and a historical trajectory of a moving target of the current frame, and a tracking obstacle list of the current frame is output. According to the method, the amount of data processing is smaller, a sensing area is wide, andcost is lower.

The invention discloses a road travel time forecasting method based on a random forest and a clustering algorithm. In the road travel time forecasting method, according to the time sequence rule of historical traffic data, combined with the road property, weather factors, holiday information and states of road upstream-and-downstream traffic flow, and through a hybrid forecasting model of the density-based clustering algorithm (DBSCAN) and the random forest (RF), travel time of all key road sections at some time interval is accurately forecasted. The forecasting result can be used for pre-judging a traffic state development tendency and making a control scheme for potential congestion roads, can also be used for dynamic path induction, can project best travel plans for travelers, and can assist in social intelligence traveling. According to the road travel time forecasting method, the forecasting accuracy of all trees in the random forest is increased through density clustering, and therefore the forecasted whole accuracy is increased.

The invention discloses a smart-phone track chain-cluster identification method considering sequential DBSCAN. The method comprises that data collection and preprocessing are carried out; the sequential angle offset of behavior track points is calculated; the sequential distance offset of the behavior track points is calculated; chains and non-chains are identified and gathered according to rules; and whether a non-chain segment belongs to a point cluster is determined by using DBSCAN, and outputs an analysis result. The logical process of the behavior track identified by the human eyes is integrated with the DBSCAN by providing the two indexes, namely the sequential angle offset and the sequential distance offset, the disadvantage that a traditional algorithm cannot identify sequential features is overcome, and the accuracy of behavior track chain-cluster identification is improved.

A system and method of identifying outliers in item categories are described. A pairwise similarity measurement may be determined between each item listing in a plurality of item listings based on a comparison of at least one feature of each item listing. At least one outlier among the plurality of item listings may be determined using the pairwise similarity measurements. The feature(s) may comprise at least one feature from a group of features consisting of: a title, an image, a price, an attribute, and a description. Each item listing in the plurality of item listings may belong to the same leaf or non-leaf category in a network-based marketplace or publication system. The outlier(s) may be determined using at least one clustering algorithm. The clustering algorithm(s) may comprise an agglomerative hierarchical clustering algorithm and / or a density-based clustering algorithm.

The invention belongs to the field of traffic road environment sensing in the intelligent automobile technology, and relates to a road edge detection method based on three-dimensional laser radar, comprising the following steps: acquiring point cloud data of the surrounding road environment by using the three-dimensional laser radar installed on the vehicle, adopting a random sampling consistencyalgorithm to segment the ground; setting a wider threshold according to various geometric features of the road edge for judgment, and extracting the candidate points of the road edge by using the neighborhood relationship between the respective scanning points of the same scanning layer; clustering the candidate points of the road edge and removing the dense and isolated noise points by using a density-based clustering algorithm according to the continuous feature about density and rod direction of the data points of the road edge; and finally fitting the qualified candidate points of the roadedge by weighted least squares to improve the fitting accuracy. The method extracts candidate points of the road edge according to the plurality of features of the road edge, and comprehensively considers the continuous feature of the road edge in the density and the road direction to de-noise, so that the final detection error is small and the precision is high.

The present disclosure relates to a method and apparatus for calculating trajectory frequency, a storage medium, and an electronic device. The method may include obtaining a set of trajectory points of a moving object; a density-based clustering partition is carried out on the track points in the track point set by using a preset neighborhood radius and a preset neighborhood track point density threshold, and a plurality of cluster sets are obtained; measurement of data center trends is performed on the plurality of cluster sets to obtain the track frequency of the moving object. The method can accurately calculate the trajectory frequency of the moving object in the frequent region, and provides a new quantitative index for the behavior analysis of the moving object and the security riskevaluation.

The invention discloses an eye movement data-based electroencephalogram experiment evaluation system and method. The method comprises the following steps of: acquiring eye movement data of an object through an eye tracker, and establishing a time-space model according to a point of regard in the eye movement data; calculating a similarity between sequences by using a dynamic time wrapping algorithm fast technology, establishing a distance matrix, carrying out outlier detection through a density-based clustering algorithm, and carrying out quantitative sorting by adoption of a learning, sorting and training model according to a clustering result so as to obtain a participation degree of the object. According to the system and method, the experiment participation degrees of objects can be objectively and quantitatively evaluated, so as to form feedbacks for experiments and models and then ensure the data quality and improve the model prediction accuracy. According to the system and method, quantitative evaluation is carried out on the experiment participation degrees of the objects, and quantitative feedbacks of emotion recognition experiment are constructed.

The invention discloses a mixed attribute data flow clustering method for automatically determining a clustering center based on density, and the method comprises the following steps: 1) initialization: carrying out the clustering of initial Ninit data objects in a data flow through a New-FSFDP algorithm, generating initial intensive micro-clusters, so as to initialize the whole on-line process and enable the mean radius of all generated initial intensive micro-clusters to serve as an initial epsilon; 2) on-line maintenance; 3) off-line clustering. The method is higher in precision, and is good in processing capability of off-group points.

The invention discloses a radar multi-target clustering method and device. The method comprises the following steps: S1, acquiring all the detected target points in a to-be-processed radar signal, anddividing all the acquired target points into multiple regions in advance according to spatial distribution density; and S2, respectively configuring corresponding input parameters used for clusteringfor each region obtained through division in the step S1, and respectively clustering each region by adopting a density-based clustering algorithm and according to the configured input parameters, thus obtaining a multi-target clustering result output; and the device comprises a region pre-division module and a multi-target clustering module. The method disclosed by the invention has the advantages of simple implementation method as well as high clustering accuracy, good clustering effect and high clustering speed when density distribution is non-uniform.

The invention discloses an image retrieval method based on BoVW (Bag of Visual Words) optimization and query expansion. The image retrieval method comprises: extracting SIFT features of a training image, and carrying out clustering on the SIFT features by utilizing a density-based clustering method to generate a BoVW set; analyzing correlation of visual words and a target category by a chi-square model, and simultaneously, filtering out a certain number of visual stop words by combining word frequencies of the visual words; carrying out mapping matching on the SIFT features and an optimized BoVW to obtain a visual word histogram; carrying out similarity matching on a visual word histogram of an query image and the visual word histogram of the training image, and according to a primary matching result, carrying out secondary or repeated retrieval by combining a query expansion strategy so as to obtain a final retrieval result. The invention provides the image retrieval method based on BoVW optimization and query expansion, which aims to improve image retrieval accuracy.

A new density based clustering method for clustering data points in multidimensional space is described. Each point has a neighborhood consisting of all points that are within a preset cutoff radius or distance. Each point is assigned a density measure based on the number of points in its neighborhood. Any point that has a higher density than any of its neighboring points is the centre of a cluster and is assigned a unique cluster ID. Every other point follows a path through the graph of neighboring points such that density is increasing as fast as possible until a cluster center is reached. The algorithm's performance is demonstrated on a one-dimensional, two-dimensional, and 18-dimensional dataset.

The invention discloses a charging method and system of electric vehicles. The method comprises the steps as follows: acquiring charged vehicle position data in the same time period; clustering the vehicle position data in the same time period with a density-based clustering algorithm to obtain position information of all charging stations; acquiring the maximum value of the number of simultaneously charged vehicles corresponding to each charging station, and determining the number of charging piles of the charging stations; acquiring the number of electric vehicles which are being charging atthe current moment in each charging station, the number of idle charging piles in each charging station and residual electricity quantity of each vehicle at the current moment; acquiring position information of a to-be-charged vehicle at the current moment; determining the position of the best charging station; sending the position of the best charging station to a user terminal so that the usercan select whether to charge the to-be-charged vehicle or not. By the adoption of the charging method and system, waiting time of drivers can be shortened and the utilization rate of the charging piles can be increased.

The present invention discloses a travel and activity mode identification method based on a BDSCAN clustering algorithm. The method comprises the following steps: cleaning traveler spatio-temporal track data sets that are continuously acquired; calculating an average speed of coordinate points of cleaned data sets, and classifying a position coordinate point whose average speed is higher than a set threshold into a travel mode; based on a DBSCAN clustering algorithm, performing clustering analysis on the cleaned data sets, and determining an activity starting point and an activity ending point according to a clustering result; and according to coordinates and time of data points of an identified travel mode and activity mode, generating a travel time table. According to the method disclosed by the present invention, based on the acquired traveler spatio-temporal track sequence sets, the behavior modes of the travelers are divided into the travel mode and the activity mode by using the density-based clustering algorithm (DBSCAN). The method disclosed by the present invention is convenient for calculation and actual operation and has strong practicality, and by the method, the behavior mode of the travelers can be determined more accurately, so as to facilitate subsequent researches, so that the method has important realistic significance.

A natural language understanding system performs automatic unsupervised clustering of dialog data from a natural language dialog application. A log parser automatically extracts structured dialog data from application logs. A dialog generalizing module generalizes the extracted dialog data to generalization identifier vectors. A data clustering module automatically clusters the dialog data based on the generalization identifier vectors using an unsupervised density-based clustering algorithm without a predefined number of clusters and without a predefined distance threshold in an iterative approach based on a hierarchical ordering of the generalization.

The present invention discloses a density-based text clustering algorithm research method. The method comprises the following steps: using the ICTCLAS word segmentation system to carry out word segmentation on a text in a text set, and extracting corresponding keywords from the word segmentation according to the three parts of speech of the noun, the verb, and the adjective, and the word frequency; using an improved HowNet word similarity algorithm to calculate keyword similarity of the obtained keywords; according to the keyword similarity in the text, calculating text similarity; and according to the obtained text similarity, using the density-based clustering algorithm to carry out clustering on the text, so that the performance of the existing text-related information retrieval technology can be significantly improved.

The invention particularly relates to a large spatial data clustering algorithm K-DBSCAN based on density. The algorithm comprises the steps that a density-based clustering parameter is preset: radius R, the minimum neighbor number Min_N, pre-division number K and division iteration number of times T are preset; a data set is divided into K1 subsets according to spatial distribution; the reachable subset of each data subset is calculated to form a reachable subset index; and based on the reachable subset index, spatial clustering based on density is carried out on the data of each subset. According to the technical scheme provided by the invention, density-based unsupervised and semi-supervised clustering can be carried out on the large spatial data set, and efficient and fast parallel clustering calculating is realized.

The invention provides a method for counting the numbers of indoor persons on the basis of WiFi (wireless fidelity) channel state information (CSI) and support vector machine (SVM) regression. Specialhardware facilities can be omitted, and the numbers of the indoor persons can be counted only by the aid of existing WiFi wireless networks; CSI data can be denoised by the aid of DBSCAN (density-based spatial clustering of application with noise) algorithms after the CSI data are acquired, then non-zero rates of each subcarrier are obtained by the aid of expansive matrix algorithms and are usedas CSI feature fingerprint samples, accordingly, the influence of great change of signal amplitudes on person number counting can be enhanced, and influence of environmental noise on small change of the signal amplitudes can be reduced; accurate nonlinear dependency relationship models between the numbers of the persons and the CSI feature fingerprint samples can be obtained by the aid of SVM regression algorithms without consideration on complicated indoor environments, and accordingly the purpose of accurately counting the numbers of the indoor persons can be achieved. The method has the advantages that the numbers of the persons can be accurately counted on the basis of the existing WiFi wireless networks, the method is low in cost and high in universality, and the privacy problems canbe solved.

A performance monitoring method for thermal power equipment based on big data analysis mining mainly comprises the following steps: step 1, implementing correlation analysis and principal component analysis on the health running data of a power station; and step 2, obtaining principal component variables of equipment by using the correlation of the equipment data and performing principal component analysis, establishing a three-dimensional cluster for normal operating conditions by performing density-based clustering analysis, and monitoring the performance of the equipment by using the cluster. Compared with a traditional performance monitoring method for the thermal power equipment, the performance monitoring method using the big data can fully use a large number of the thermal power plant operating data, and thus the real-time performance of the equipment can be reflected, and the monitoring accuracy can be increased.

The invention discloses a field crop three-dimensional reconstruction method based on laser radar point cloud, and the method comprises the steps: obtaining the point cloud data of a field crop through a laser radar, and carrying out the data preprocessing of the point cloud data; extracting a small amount of point clouds as key control points, and establishing a digital surface model and a digital ground model by utilizing the key control points and through an irregular triangulation network; constructing an initial spike layer space model according to the digital surface model, and obtainingpoint cloud data in an actual spike layer space model through an Otsu threshold method; segmenting the point cloud in the actual spike layer space model into a plurality of point cloud clusters by adopting a density-based clustering algorithm; correcting the number of point cloud clusters, the number of the point cloud clusters being the number of crop ears, and calculating the crop height according to a digital ground model; and constructing a three-dimensional model of the crops in the field according to the obtained crop number, crop height and crop spike coordinates. According to the invention, the construction of the three-dimensional model of the field crops is realized, and enough data support is provided for the digital management of the field.

The invention discloses an HTAP-oriented distributed database intelligent hybrid storage method. The method comprises the steps that data in a data source are obtained through a data import system andstored in a storage engine of an HTAP database, the storage engine is composed of a plurality of storage nodes, and the data of each storage node are stored in the data organization format of a column family; and the central node performs optimized recombination analysis on the data layout in the storage engine by adopting a density-based clustering partition algorithm according to the working load conditions of the historical OLAP and OLTP, so as to obtain the optimal data layout. According to the invention, the data in the HTAP database is organized in a column family manner; and the data in the column group is dynamically adjusted according to the optimal storage layout calculated by the central node, and the optimal storage layout is obtained by a density-based clustering algorithm through a clustering result, so that the columns with equivalent access frequencies belong to the same column group, i.e., the columns which are frequently accessed belong to the same column group.

The invention discloses a wind power plant and power-to-gas plant and station collaborative location planning method. The method comprises the following steps: 1) establishing a mathematic model of a power-to-gas plant and station; 2) with historical wind speed time series being an original scene, carrying out clustering analysis on a large number of historical data by utilizing a density-based clustering algorithm to obtain a reduced wind power plant output scene; 3) according to the mathematic model obtained in the step 1) and the wind power plant output scene obtained in the step 2), and with net investment income maximization being an optimization objective, establishing a collaborative location planning mathematical model based on scene analysis; and 4) according to the steps 1)-3), solving the constructed collaborative location planning mathematical model of the wind power plant and the power-to-gas plant through an AMPL / BONMIN solver to obtain a planning result. The collaborative location planning method can take technical features of operation of the power-to-gas plant and station into consideration, can evaluate and analyze economy of the planning scheme and can effectively reduce system wind abandoning quantity.

Embodiments are directed to a density-based clustering algorithm that decomposes and reformulates the DBSCAN algorithm to facilitate its performance on the Map-Reduce model. The DBSCAN algorithm is reformulated into connectivity problem using a density filter method and a partial connectivity detector. The density-based clustering algorithm uses message passing and edge adding to increase the speed of result merging, it also uses message mining techniques to further decrease the number of iterations to process the input graph. The algorithm is scalable, and can be accelerated by using more machines in a distributed computer network implementing the Map-Reduce program.

The invention discloses a method for identifying modal parameters of an ocean platform structure. The method comprises the following steps of 1, carrying out directional processing on input environmental excitation response data of the ocean platform structure; 2, constructing a Hankel matrix by utilizing the structural environment excitation response data, carrying out orthogonal triangular decomposition on the Hankel matrix to obtain a projection matrix, then carrying out singular value decomposition on the projection matrix, and solving a vibration system state space equation by utilizing aleast square method and the like; and finally, extracting structural modal information from the system matrix A and the output matrix C in the state-space equation of the vibration system; 3, performing clustering analysis based on density division on the modal result set by utilizing a density-based clustering algorithm; and 4, obtaining real modal parameters, and analyzing a result of the realmodal parameters. According to the method, the real mode of the ocean platform structure can be automatically extracted from the environmental response signal. Therefore, most of noise points are eliminated. Moreover, the health condition of the platform can be evaluated according to a modal parameter identification result.