225 results about "DBSCAN" patented technology

The invention discloses a road environment element sensing method based on a laser radar, and the method comprises the steps: 1, downsampling original point cloud data of the laser radar, obtaining aregion of interest, obtaining simplified point cloud data, segmenting ground data, and obtaining a result; 2, adopting a voxelization-based DBSCAN clustering algorithm to realize barrier segmentationof non-ground data, and carrying out obstacle classification and recognition based on multiple features; and 3, designing a road boundary extraction method suitable for vehicle-mounted laser radar data by utilizing data point elevation mutation, screening boundary points by using prior knowledge after extracting the road boundary points, fitting a road boundary line by adopting a least square method, and marking a barrier-free area by using the grids to realize detection of a passable area in a laser radar road scene. The method is simple in algorithm, small in data error, safe and reliable.

The invention provides an acquisition method and apparatus of the operation area of farm machinery. The acquisition method of the operation area of farm machinery includes the steps: receiving the farm machinery operation track data sent from a farm machinery positioning device; based on the operation speed, improving the neighborhood radius determination method of the dbscan clustering algorithm; utilizing the improved dbscan clustering algorithm to filter the road driving point and the field transition point in the farm machinery operation track data; according to the filtered farm machinery operation track data, determining the number of fields operated by the farm machinery; and utilizing the distance method to calculate the area of each field operated by the farm machinery. The acquisition method and apparatus of the operation area of farm machinery can adaptively determine the neighborhood radius according to the operation speed of farm machinery, can effectively segment the adjacent fields operated by the farm machinery, and can improve the accuracy and automation degree for acquisition of the operation area of the farm machinery.

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 a driver fixation point clustering method based on a density clustering method and a morphology clustering method, and belongs to the field of typical density clustering methods and mathematic morphology clustering methods. The driver fixation point clustering method includes the steps of putting forward a density method and mathematic morphology method combined self-adaption DBSCAN-MMC method, applying the method to driver fixation point clustering, setting the value of the Eps through fixation point structure parameters, obtaining an initial point set of MMC clusters through the DBSCAN, determining the number of the clusters, reducing outliers produced through DBSCAN clusters through the self-adaption MMC clusters, and completing clustering oriented to driver fixation areas. According to the method, the advantages of irregular shape clustering of the DBSCAN and the MMC are fully used, the defects of the two clustering methods are overcome, the clustering effect is superior to the clustering effect of the conventional DBSCAN clustering method and the conventional MMC clustering method when the driver fixation areas are divided, and the driver fixation clustering quality is improved.

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 relates to a layered point cloud segmentation method based on DBSCAN. Firstly, a CSF is adopted to separate a ground point from a non-ground point; In a non-ground point segmentation process, point clouds in the vertical direction are layered according to a certain height, DBSCAN clustering is carried out on projection points of each layer on an XOY plane to obtain a central point ofeach cluster, all the clustered central points are projected to the XOY plane, and each object main body is clustered by utilizing DBSCAN to obtain a plurality of object main bodies; a judgment is made whether each main body point exists in each layer of each main body or not, judging the number of objects contained in each cluster, and finally, a cluster of multiple objects is segmented. For segmentation of side viewpoint cloud data, extraction of most of main bodies in a scene can be guaranteed, certain robustness is achieved, particularly, the invention has good performance in the scene with trees as the main component, and the result obtained through the method has certain significance in point cloud classification and point cloud three-dimensional reconstruction after point cloud segmentation.

The invention discloses a millimeter-wave radar environment map construction method and device based on a clustering algorithm. The method comprises the steps that: multi-frame mobile acquisition dataand corresponding pose information of a millimeter-wave radar are acquired; according to the pose information, the coordinate information of the target of the multi-frame mobile acquisition data relative to the radar in a Cartesian coordinate system is processed to obtain combined point cloud data, and the combined point cloud data comprises a plurality of sample vectors; a DBSCAN clustering algorithm is used for carrying out miscellaneous point screening processing on the combined point cloud data, and input parameters of the DBSCAN clustering algorithm are determined according to dimensioninformation of the combined point cloud data and Mahalanobis distance data between sample vectors in the combined point cloud data; a millimeter-wave radar measurement precision model is established by utilizing the combined point cloud data subjected to miscellaneous point screening processing; and a millimeter-wave radar environment map is constructed according to the millimeter-wave radar measurement precision model. The method can effectively improve the accuracy of the millimeter-wave radar environment map.

The embodiment of the invention provides a checked-in hotel surrounding life recommending system based on a machine learning statistical model. The system is characterized by comprising an information acquisition module for acquiring user data, corresponding related information of urban geological positions and POI (point of interest) data of a map APP (application); a data storage module for storing data based on class in formatted manner, storing data acquired by the information acquisition module, and data calculated via an algorithm module; the algorithm module for calculating and training data via the algorithm DBScan; a matching strategy module for executing a user-residence and user-city POI matching strategy, wherein the matching strategy performs different strategies or a combination of the strategies according to service logic, user interest or user-fed city weather; a recommending engine module for displaying to a user, matched POI coordinate data. The invention also provides a checked-in hotel surrounding life recommending method based on the machine learning statistical model.

The invention discloses a C-DBSCAN-K clustering algorithm under a Hadoop platform. The algorithm comprises the following steps of step 1.establishing clusters capable of communicating with each other; step 2.establishing the Hadoop platform for the clusters; step 3.using a dfs-put command to upload a to-be-clustered data set A to an HDFS; step 4.executing a Canopy clustering algorithm to carry out initial clustering on data in the A in order to obtain a clustering result of coarse granularity; step 5.constructing a k-d tree on the clusters obtained in the step 4; step 6.executing a DBSCAN algorithm to the clusters obtained in the step 4, using the k-d tree to query an epsilon-neighborhood of a data object in each cluster and outputting a clustering result; and step 7.merging the clusters with the same data in the step 6 and outputting a clustering result. The algorithm of the invention solves a problem of low clustering efficiency of the DBSCAN clustering algorithm on a large-scale data set in the existing technology.

The invention discloses an abnormal electric quantity data identification method based on limit value learning. The method comprises the following steps: analyzing abnormal electric quantity data to obtain the type and identification algorithm of the abnormal electric quantity data; then, analyzing and researching an OnClassSVM algorithm; learning the identification limit value of the abnormal electric quantity data to obtain a limit value learning table, checking the abnormal electric quantity data of the historical data through the limit value learning table, and then checking an outlier inthe historical data through analyzing a density-based clustering algorithm DBSCAN algorithm to realize abnormal electric quantity data identification based on limit value learning; finally, analyzingand researching the same density-based clustering algorithm LOF algorithm, combining the two density-based clustering algorithms to carry out an experiment, and carrying out outlier identification onthe multi-dimensional data to realize the multi-dimensional electric quantity data outlier identification based on the density clustering algorithm.

The invention discloses a grouping method in a wind power plant based on extreme gradient dynamic density clustering, and the method comprises the steps: selecting indexes of grouping in the wind power plant, and carrying out the abnormal value detection and truncation of corresponding index data in a certain time period; for the preprocessed grouping index data, carrying out dimension reduction selection on the grouping index data by adopting XGBoost; and carrying out cluster division on the selected index data on the basis of DBSCAN-DTW clustering. According to the method, the problem of partial missing of actual wind power plant data can be effectively solved, and the model accuracy is improved; and the method is used for processing the multi-dimensional time sequence characteristic operation data of the fan, so that accurate and effective division of the groups in the wind power plant can be obtained.

The invention belongs to the technical field of electronic countermeasures, and particularly relates to a passive multi-station multi-target positioning method based on DBSCAN. The method provided bythe invention mainly comprises: firstly, determining an observation line of sight by combining angle information relative to a target measured an observation station within the own position information of the observation station, then obtaining the location of an intersection of all lines of sight according to the observation line of sight, finding out a cluster formed by the intersections in thevicinity of various targets through a density-based DBSCAN clustering method, and finally realizing multi-target positioning through the central location of the found cluster. The passive multi-station multi-target positioning method provided by the invention has the beneficial effects that positioning estimation can be effectively performed on multiple targets through the angle information measured by multiple observation stations and the own position of the observation stations, the method is simple, and the effect is good.

The invention relates to a modulation format recognition and OSNR estimation combined method for density clustering, and belongs to the technical field of modulation recognition in coherent light communication. The method comprises the following steps: constructing a time domain constellation diagram by using two paths of signals of coherent received data, identifying data clustering clusters meeting a preset density threshold based on a density clustering method of DBSCAN, and dividing a judgment threshold to identify a modulation format of the received data by using the number of the clustering clusters as a factor of modulation format identification; and calculating the proportion of the data meeting the density threshold in the total data, and performing high-order polynomial fitting by utilizing the density information to realize OSNR estimation. Accurate identification and OSNR low-error estimation can be performed on QPSK, 8PSK, 16QAM, 32QAM and 64QAM without a large amount of training data. The method has the potential of low system complexity and real-time monitoring of signals, and has application prospects in bit error rate calculation and linear and nonlinear damage monitoring.

The invention discloses a moving target detection method, an electronic device and a storage medium. The moving target detection method, the electronic device and the storage medium are applied to thetechnical field of radar signals. An image domain moving target detection method is provided for a single-channel satellite-borne synthetic aperture radar (SAR) bunching mode, the satellite-borne SARlong-term observation and the characteristics of movement of a moving target signal in a sub-aperture image sequence are fully utilized, the speed of the moving target detection method is higher thanthat of an existing single-channel iterative detection algorithm, the position of a moving target on an SAR image can be automatically acquired, through the clutter image difference, a DBSCAN and a Kalman tracking algorithm are improved to effectively reduce false alarms, no raw data is required, only single look complex images are input, and no raw SAR echo data is required.

The invention discloses a multi-granularity urban public bicycle scheduling method and a system based on DBScan. The scheduling method is implemented based on the following clustering method: S110, using a site as a clustering object to draw a k-dist graph of the clustering object; S120, extracting an inflection point of a k-dist graph curve, wherein region radius corresponding to all inflection points form a neighborhood radius value set; and S130, clustering the clustering objects of the corresponding levels by adopting the DBScan method for each neighborhood radius value, obtaining clustersof different levels. According to the clustering result, the multi-granularity urban public bicycle scheduling method and the system based on DBScan respectively obtains the regions formed by all theclusters corresponding to each level that is a scheduling unit. According to the hierarchy from high to low, a city public bicycle scheduling scheme of each level is sequentially formulated. According to the multi-granularity urban public bicycle scheduling method and the system based on DBScan, the problem of poor clustering quality caused by uneven distribution of public bicycle stations can effectively solved, and the invention benefits to improving the clustering quality and furthermore improving the effectiveness of scheduling.

The invention relates to a telecommunication fraud gang clustering method and a telecommunication fraud gang clustering system based on a DBSCAN algorithm. The method comprises the steps of obtainingthe suspicious communication data; analyzing called party data in the suspicious communication data to obtain calling party data corresponding to the called party data; taking the collected data of each calling party as a telecommunication network node, and forming a telecommunication network node set in a predetermined area; judging whether each telecommunication network node in the predeterminedarea is a core point or not so as to find out all the core points in the predetermined area; finding out boundary points and noise points in a predetermined area according to each core point; and deleting the noise points, and forming a telecommunication fraud gang cluster according to all core points and boundary points in the predetermined area. Based on the DBSCAN algorithm, the telecommunication fraud gang cluster existing in any shape can be found, the whole telecommunication fraud gang can be eradicated at a time, and the real-time and effective fraud early warning and the low-cost treatment are achieved.

The invention discloses an active memory prediction migration method in a virtual machine migration process and belongs to the technical field of virtual machine migration. The method comprises the following steps of: firstly, preprocessing a pre-recorded page set consisting of an address of an accessed memory and accessed time recorded by a virtual machine manager by adopting a method of expressing each memory page by using a six-tuple; and then calculating a priority weight of the memory page, and performing clustering analysis by adopting an ISS-DBSCAN clustering algorithm to determine an active memory range; and finally, adjusting the sending sequence of the active memory pages by taking the priority weight of the memory pages as a judgment basis of the priority. According to the method, a theoretical basis and a method are provided for active memory analysis, the activity degree of the memory page can be effectively predicted, and the memory migration efficiency is improved.

The invention discloses a traffic target identification method based on a DBSCAN algorithm, and belongs to the technical field of data processing. The method comprises the steps of: firstly, using a millimeter wave radar for detecting a to-be-detected target in a continuous time period, and obtaining different position information of the to-be-detected target; secondly, taking the position information as point cloud data, and clustering the point cloud data by using a DBSCAN clustering algorithm to obtain clusters; performing identification and division of target types by using the number of scattering points in the clusters; after the target types corresponding to the clusters are obtained, counting the number of the target types, and finally completing identification and counting of traffic targets in comprehensive traffic environment. The target identification accuracy is improved, and the identification process is simple and efficient.

The invention belongs to the technical field of electronic countermeasure, and particularly relates to a DBSCAN-based passive MIMO multi-target positioning method by using a time delay measurement value. According to the method, a multi-target data association algorithm is mainly adopted, time delay information measured by a plurality of receiving stations is converted into distance sums, and association is carried out to obtain a cost matrix C (k, j). Firstly, data preprocessing is carried out, data larger than a threshold value in a cost matrix are removed, then cluster division is carried out through a DBSCAN clustering algorithm, and finally weighted fusion is carried out on different clusters to obtain position estimation of multiple targets. The method has the beneficial effects thatthe MIMO multi-target data can be accurately associated, the position of the target can be finally and accurately estimated, the method is simple, and the effect is good.

The invention discloses a power equipment state monitoring data clustering method and system, and the method comprises the steps: obtaining a to-be-clustered data set Xm * n which comprises m samples,enabling each sample to comprise n types of variables, and standardizing the n types of variables of each sample; setting a density threshold parameter k of the DBSCAN model; drawing a k distance mapaccording to the distance between each sample in the standardized Xm * n and other samples; determining a lower limit threshold E0 of the neighborhood radius parameter E according to the category number of the clustering result of a DBSCAN (0, k) model; determining an upper limit threshold Emax of E according to the k distance graph; drawing a'category number-E 'curve according to the E0 and theEmax; and determining the optimal value of E according to the'category number-E 'curve. Optimization of a neighborhood radius parameter E and a density threshold parameter k of a DBSCAN model is achieved by designing and drawing a category number-E curve, and the DBSCAN model is used for clustering power equipment online monitoring data and used for pattern recognition and anomaly detection of online monitoring real-time data and judging normal data categories and anomaly types.

The invention belongs to the technical field of IT operation and maintenance and machine learning, and particularly relates to a method for realizing fault detection by using a time sequence clustering algorithm. The method comprises the following steps of acquiring the equipment performance index information according to a preset time frequency to obtain the time sequence data; normalizing the time series data; performing clustering analysis on the normalized time series data by using a DBSCAN algorithm, and calculating an abnormal value score of the clustered time series; and judging whethera fault exists or not according to whether the abnormal value score exceeds a set threshold value or not. According to the method for realizing the fault detection by using the sequential clusteringalgorithm, the DBSCAN algorithm is used for carrying out clustering analysis on the equipment time sequence data, and whether the equipment performance state is stable or not is judged by analyzing the difference value between all performance data indexes, so that the equipment operation health degree is measured, and the detection efficiency and accuracy can be effectively improved.

Provided is a self-adaption landmark selection method facing moon navigation. The method comprises the following steps: firstly, extraction of sift characteristic points is carried out by utilization of the SiftGPU algorithm; secondly, downsampling of characteristic points is carried out; thirdly, the characteristic points in the second step is subjected to clustering by utilization of the self-adaption DBSCAN cluster algorithm, and the process is as follows: firstly, distribution is carried out according to a shortest distance of each characteristic point, and the initialized parameter of the cluster algorithm is obtained and secondly, a non-recursive mode is employed to achieve the DBSCAN algorithm, and a plurality of candidate landmarks are obtained; fourthly, a characteristic point M in a current landmark in correct matching, Mmax with the most characteristic points in all the landmarks in matching and detected characteristic point number A are obtained through matching of two adjacent images, and the landmark with the highest score is selected as a landmark by utilization of an evaluation function. The provided method is advantaged by good self-adaption capability and good real-time.

The invention relates to a backdoor confrontation sample generation method of a PE malicious software detection model based on R-DBSCAN, and belongs to the field of computer malicious software detection. The method mainly aims at solving the problem that a malicious software detection model is high in attack difficulty under the black box condition. The method comprises the following steps: firstly, acquiring a PE sample from a public data set, training a proxy training model, and reducing the dimension of the data set by adopting an SHAP value; clustering the samples by adopting an R-DBSCAN method, and taking a center node of each cluster as a sampling point to construct a new data set; training a neural network model; respectively inputting malicious and benign sample files, and recording neurons which greatly influence a classification result according to the weight change condition of the neurons in the neural network; embedding a character string with any length into the empty PE file, taking a character string which greatly influences the character string according to the weight change condition of the neuron, and recording the neuron; embedding a trigger into an original malicious PE file, and modifying a label to achieve adversarial training of a neural network.

The invention discloses a moving trajectory big data-oriented optimal waiting point recommendation method and system. The method comprises the following steps: S1, obtaining moving trajectory data ofa vehicle; S2, preprocessing the movement track data to obtain vehicle getting-on and getting-off hotspot data; S3, constructing a parallel SP-DBSCAN algorithm according to the get-on and get-off hotspot data; S4, performing clustering analysis by using an SP-DBSCAN algorithm to obtain a plurality of waiting point recommendation areas; S5, according to the plurality of waiting point recommendationareas, obtaining a plurality of centroids of each waiting point recommendation area; S6, recommending an optimal waiting point according to the plurality of centroids and the positions of the passengers, constructing a parallel SP-DBSCAN algorithm, performing clustering analysis by using the SP-DBSCAN algorithm to obtain a waiting point recommendation area, and obtaining the optimal waiting pointaccording to the waiting point recommendation area; therefore, the technical problems of distributed storage and parallel computing of optimal waiting point recommendation based on moving track big data are solved, and the efficiency of processing large-scale moving track data is high.

The invention discloses an intelligent scheduling platform for customer visit. Clustering is performed according to the distance, scheduling optimization is carried out on the AOI data; one shopping mall cannot complete sheduling in one day at most; the number of days is calculated according to the quantity, a VRP model is adopted for distribution; a dbscan clustering algorithm is used in the clustering process; the DBSCAN is a clustering algorithm based on a density space, can be used for concave data sets, is suitable for clustering analysis of irregularly distributed client points; automatic customer return visit is achieved by utilizing spatio-temporal data and based on a machine learning algorithm and corresponding constraint conditions; engineering and automation of the return visitschedule and path planning are achieved, the return visit schedule arrangement can be automatically completed as long as an operator updates client data and corresponding constraint rules, the accuracy can be as high as an hour level, the effectiveness of client return visit can be greatly improved, the operation cost is saved, the operation efficiency is improved, and the satisfaction degree of clients is improved.

The invention discloses a wind power plant aggregation equivalent modeling method based on principal component analysis and a clustering algorithm, which aims at the characteristics of complex operation state and high data dimension of a wind turbine generator, considers the influence of variable correlation and data redundancy on a grouping effect, and adopts the principal component analysis and the DBSCAN clustering algorithm to group a wind power plant. According to the method, grouping indexes are selected according to unit types to extract operation data, principal component analysis is applied to carry out dimension reduction processing on the data, a DBSCAN algorithm is applied to carry out unit group division, a capacity weighting method is adopted to calculate equivalent unit parameters, equivalent wind speed is calculated with the principle that input wind energy is equal, and current collection line equivalent parameters are calculated according to an equal power loss method. The established equivalent model can accurately represent the operation characteristics of the wind power plant and is used for analyzing a grid-connected simulation result, and the method plays a very important role in researching the characteristics of a large-scale wind power plant access power system, reducing the requirements on simulation hardware performance, saving simulation time and maintaining the safe and stable operation of the power system.

The invention discloses a Chinese character field detection method and system based on character recognition, and the method comprises the steps: recognizing a character region in a pre-selected region, collecting the character region, and calculating a relative distance matrix; clustering the matrix based on DBSCAN to obtain a character string region; extracting characters from the character string area through a sliding window and then putting the characters into a single-character classifier, and obtaining a predicted Chinese character field; training a single character classifier forwardsand backwards through CTCLoss on the basis of the predicted Chinese character field, and outputting the character probability through a softmax function; and putting the characters extracted by the sliding window into the trained single-character classifier to obtain a Chinese character field. Through the relative distance matrix word clustering and forward and reverse CTCLoss training of a sliding window classifier, Chinese character fields in a complex environment can be accurately identified. The problems of inaccurate character string extraction and overlarge model time and space complexity are solved, and the Chinese character field in a complex environment can be identified more accurately.

The invention is suitable for the field of computers, and provides a photovoltaic power abnormal data recognition method and device, and terminal equipment. The method comprises the steps: obtaining the photovoltaic power of a photovoltaic power station at different time, and obtaining a time sequence power data set; clustering the time sequence power data set by adopting a Kmeans clustering algorithm to obtain a clustering data set; based on the clustering data set, calculating the deviation of the data points and the clustering centers corresponding to the data points to obtain a deviation data set; clustering the deviation data set by adopting DBSCAN to obtain an abnormal data distance threshold value; and based on the clustering center and the abnormal data distance threshold, classifying the deviation data set to obtain an abnormal data set. According to the method, the photovoltaic power abnormal data is recognized through the Kmeans and DBSCAN second-order clustering algorithm,and the flexibility and the adaptability of abnormal recognition are improved by utilizing data features in the global dimension.

The invention discloses a precision marketing method based on Xgboost and DBSCAN. The method comprises the following steps: 1, carrying out the manual marking of bank data in a data source, and obtaining the class information; 2, integrating the bank data and the annotation information of the bank data to obtain a complete data set; 3, exploratory analysis is conducted on the data set; 4, performing data preprocessing on the data set, and dividing the data set to obtain a training set and a test set; step 5, inputting the data into a constructed model for training to obtain a precision marketing model; 6, inputting to-be-detected bank data into the precision marketing model for detection to obtain a customer classification result; and step 7, performing effect verification, and outputtinga detection result. According to the invention, the target client is found through Xgboost, and the DBSCAN is used to recommend the corresponding product to the target client, so that the marketing success rate is improved.