224 results about "Spectral clustering algorithm" patented technology

The invention discloses a SAR (Synthetic Aperture Radar) image segmentation technique based on dictionary learning and sparse representation, and mainly solves the problems that the existing feature extraction needs a lot of time and some defects exist in the distance measurement. The method comprises the following steps: (1) inputting an image to be segmented, and determining a segmentation class number k; (2) extracting a p*p window for each pixel point of the image to be segmented so as to obtain a test sample set, and randomly selecting a small amount of samples from the test sample set to obtain a training sample set; (3) extracting wavelet features of the training sample set; (4) dividing the training sample set by using a spectral clustering algorithm; (5) training a dictionary by using a K-SVD (Kernel Singular Value Decomposition) algorithm for each class of training samples; (6) solving sparse representation vectors of the test sample on the dictionary; (7) calculating a reconstructed error function of the test sample; and (8) calculating a test sample label according to the reconstructed error function to obtain the image segmentation result. The invention has the advantages of high segmentation speed and favorable effect; and the technique can be further used for automatic target identification of SAR images.

The invention relates to a gait classification method based on multi-sensor information fusion. The gait classification method includes the following steps that (1), plantar pressure information and ankle angle information in the walking process of multiple patients are collected; (2) according to the obtained plantar pressure information, gait stages of the patients are analyzed, the gait stages are divided into the stages of touching the ground with the feet and the stages of swinging the legs, and one gait cycle includes one stage of touching the ground with one foot and one stage of swinging the leg of the same foot of each patient; (3) a characteristic value of each gait cycle is set, so that gait characteristics of all the patients in the walking process are represented; (4), gait cluster analysis is performed on all the characteristic values of all the gait cycles of all the patients through a spectral clustering algorithm, so that the patients with the different gait characteristics are divided into different classifications. The patients are objectively classified, so that reference is provided for rehabilitation training and treatment of the patients, and a doctor can adopt different treatment modes and training intensities for the patients of the different classifications conveniently. The gait classification method can be widely used in the fields of gait analysis and medical rehabilitation.

The invention provides a recording device clustering method based on Gaussian mean super vectors and spectral clustering. The method comprises the steps that the Melch frequency cepstrum coefficient MFCC characteristic which characterizes the recording device characteristic is extracted from a speech sample; the MFCC characteristics of all speech samples are used as input, and a common background model UBM is trained through an expectation maximization EM algorithm; the MFCC characteristic of each speech sample is used as input, and UBM parameters are updated through a maximum posteriori probability MAP algorithm to acquire the Gaussian mixture model GMM of each speech sample; the mean vector of all Gaussian components of each GMM is spliced in turn to form a Gaussian mean super vector; a spectral clustering algorithm is used to cluster the Gaussian mean super vectors of all speech samples; the number of recording devices is estimated; and the speech samples of the same recording device are merged. According to the invention, the speech samples collected by the same recording device can be found out without knowing the prior knowledge of the type, the number and the like of the recording devices, and the application scope of the method is wide.

The invention discloses a conference recorder with a speaker speech extracting function. The conference recorder with the speaker speech extracting function comprises a main control module, a recording and playback module, a removable storage module, an interaction and display module and a speaker speech processing module, wherein the speaker speech processing module comprises a speaker segmenting module and a speaker clustering module. The main control module transmits a conference speech stream to the speaker segmenting module, and the speaker segmenting module detects speaker changing points in the speech stream and segments the speech stream into a plurality of speech sections according to the changing points; the speaker clustering module carries out speaker clustering on the segmented speech sections through a spectral clustering method, the speech sections of the same speakers are jointed together in sequence, and thus the number of the speakers and the speech of each speaker are obtained. The conference recorder and the speech extracting method are capable of automatically extracting the speech of each speaker from the conference speech, comprehensive in function and convenient to use.

The invention relates to a terminal area prevailing traffic flow recognizing method based on track spectral clusters. The method includes the steps of firstly, analyzing a given airport pavement to obtain flight path data, and conducting dividing to obtain feature track points and normal track points; secondly, setting up a space rectangular coordinate system; thirdly, putting forward an occupation degree concept according to the distance between a track and the center of a space grid, and enabling the occupation degree concept to be used for representing the occupation degree of the track in the space grid; fourthly, setting up an inter-track overall similarity model on the basis of a track partial similarity model; fifthly, constructing an Laplacian similarity matrix, and then analyzing the clusters through the spectral cluster algorithm; sixthly, conducting prevailing traffic flow recognition on the clusters obtained in the fifth step through the nuclear density estimation method; seventhly, displaying the recognition result in a displaying and interaction module. The terminal area prevailing traffic flow recognizing method has the advantages that a prevailing traffic flow track and an abnormal track can be simultaneously obtained through the spectral cluster algorithm, therefore, related personnel are assisted in scientifically and reasonably planning a terminal area and improving airport entering and leaving air lines, and the capacity of the terminal area is improved.

An image classification method based on a deep learning feature and a maximum confidence path belongs to the field of mode identification. The method comprises the steps of: training a convolutional neural network on a large enough image library; extracting an image feature by means of the trained convolutional neural network; calculating a mean vector of each class; performing iteration clustering on the mean vector that represents each class by means of a spectrum clustering algorithm so as to construct a visual tree; training svm for each non-leaf node of the tree; and for a given test image, from top to bottom, judging the probability of the test image to the corresponding node, and finding a leaf node with the biggest path probability, namely a final target class. The image feature is extracted by means of CNN, thereby achieving very good discrimination and robustness; a distance calculation formula of two classes is given out, the complexity of calculation is greatly optimized by means of derivation and the similarity of the classes is obtained, so that the visual tree is constructed by iteratively using the spectrum clustering algorithm; and the use of a visual relationship between the classes can achieve very good effects for large image classification.

The invention discloses a partial model weight fusion Top-N film recommending method based on user clustering. The method comprises the steps of 1, preprocessing data, wherein inactive users and filmswith very low popularity are subjected to data cleaning, user film label documents are constructed, explicit scoring information is converted into implicit feedback information, and a user-film implicit feedback matrix A is constructed; 2, conducting user clustering, wherein film label information is utilized, user feature vectors are obtained by training an LDA topic model, and user clustering is achieved through a spectral clustering algorithm; 3, determining a local recommending model and training a global recommending model; 4, conducting model weight fusion recommending; 5, proving the effectiveness of the models through a leave-one-out method.

The invention relates to an enterprise electricity consumption analysis and prediction method based on data mining. Multiple influence factors, including temperature, moisture, holidays and the like,are combined to analyze and predict the enterprise electricity consumption. The method comprises the following steps that: firstly, utilizing a Newton interpolation method, a normalization method anda PAA (Piecewise Aggregate Approximation) algorithm are used for carrying out preprocessing; then, utilizing a spectral clustering algorithm to carry out clustering on a dataset, judging and correcting an abnormal data to obtain enterprise electricity consumption groups, including temperature, moisture, holidays and the like with high correlation; and finally, selecting the same class of enterprise electricity consumption data and the influence factors with high correlation as the prediction input of the model, and utilizing an RNN (Recurrent Neural Network) to obtain a prediction value. By use of the method, according to different enterprise electricity consumption types, an electricity consumption influence factor is combined to construct different prediction models, and therefore, the effects of high model prediction accuracy and data preprocessing ability can be achieved.

The invention relates to a load aggregate grouping forecasting method based on a gated circulation unit network, comprising the following steps: clustering user load data by using an adaptive distributed spectrum clustering algorithm, thereby obtaining a plurality of power consumption groups with similar load characteristics and obtaining load characteristic matrices of each group; Three GRU networks are constructed, and three GRU networks are trained by extracting the temporal characteristics of the population to obtain three GRU network forecasting models, and three GRU networks are fused bystochastic forest algorithm to obtain the load forecasting models of each population. The characteristics of the time to be forecasted are inputted into the load forecasting model, the load forecasting values of each group are obtained respectively, and the forecasted values of different groups are summed up to obtain the forecasted values of the final load aggregates. By introducing grouping prediction, depth neural network and model fusion method, the invention can grasp user load characteristic and variation rule, and has high prediction precision and strong applicability.

The invention discloses a wave band selection method for a hyperspectral remote-sensing image. The wave band selection method improves the traditional wave band selection method for the hyperspectral remote-sensing image, which is analyzed on the basis of the important point of a time sequence. The wave band selection method for the hyperspectral remote-sensing image comprises the following steps: firstly, on the basis of visual clustering tendency evaluation, clustering by a spectral clustering algorithm to reduce a clustering number search range and improve the clustering quality; then, when an important redundant wave band is finally reduced, removing parts of a high-redundancy wave band according to the condition mutual information among wave bands; and searching an optimal wave band combination with a branch and bound method to improve the classification precision and reduce a final wave band number. Compared with the prior art, the wave band selection method for the hyperspectral remote-sensing image, which is disclosed by the invention, has a bigger advantage on the aspects of finally-selected wave band numbers and the corresponding classification correction rate, and the required calculation time is far lower than calculation time required with most traditional methods.

The invention discloses a terminal sector dividing method based on the graph theory and the spectral clustering algorithm. The method is carried out with the help of a computer system which comprises a sector dividing subsystem. The terminal sector dividing method comprises the first step of putting forward a calculating model of a peak connection degree according to the basic airline network structure and air traffic flow in controlled airspace, the second step of effectively dividing peaks of an airspace image based on the spectral clustering algorithm to solve the problem of a division error in short-distance parallel air routes and to achieve construction and division of a sector convex hull, the third step of putting forward a boundary optimized path selection algorithm based on sectors of an MAKLINK image, and the last step of carrying out optimizing to eliminate a sawtooth shape of the boundaries of the sectors to some degree to enable the sector boundaries to be more conform to actual operations so that the final dividing of the sectors can be finished. By means of air traffic flow, reverse deduction is carried out conveniently while the sectors are divided. By means of the terminal sector dividing method based on the graph theory and the spectral clustering algorithm, flows of the sectors are balanced, coordination is small, minimum distance constraint is met, the difficulty of commanding of a controller is reduced, and the security of operation in the terminal area is guaranteed.

The invention discloses a traffic accident risk estimation method and system. According to the method and the system, a corresponding relation chain between a risk factor and a risk class is calculated and determined by acquiring an expressway tunnel traffic accident risk factor and an expressway tunnel traffic accident risk class and employing a spectral clustering algorithm; a risk estimation model is constructed according to the relation chain and an extreme learning algorithm; and the final risk class corresponding to the risk factor can be obtained by inputting arbitrary risk factor data to the risk estimation model. Therefore, with the adoption of the method or the system provided by the invention, the expressway tunnel traffic accident risk estimation efficiency is effectively improved, so that the expressway tunnel traffic accident risk estimation efficiency process is more high-efficiency and convenient, and the safety evaluation of expressway tunnel operation period is accurately and effectively achieved.

A method for clustering large-capacity text includes presenting vector space of text, calculating similarity between each two texts, embedding pattern into dimensional space and using K-means algorithm to cluster texts to be two types, carrying out successive bisect till requirement is satisfied and pattern is not divided any more.

The invention provides a subspace clustering method based on high-dimensional overlapping data analysis and relates to the technical field of machine learning. The method builds a weighted mixed normsubspace representation model for a data matrix that needs clustering; obtaining an optimized coefficient matrix in the weighted mixed norm subspace representation model by using a linear alternatingdirection method; establishing a similarity matrix based on the optimized coefficient matrix; dividing the similarity matrix into subspaces by using a spectral clustering algorithm to obtain an initial clustering results; establishing an overlapping probability model of the subspaces; applying the overlapping probability model to an initial subspace division result to determine the overlapping ofthe subspaces; verifying a subspace clustering result to obtain a final clustering result. The subspace clustering method based on high-dimensional overlapping data analysis can improve the density ofthe same subspace data and the sparsity of different subspace data, and improve the accuracy of clustering.

The invention is applicable to the technical field of privacy protection, and provides a spectral clustering method based on differential privacy protection. The method includes the steps of pre-processing sample data; calculating a similarity matrix; based on a k-near-value, simplifying the similarity matrix; adding the random noise satisfying Laplace distribution to the similarity matrix; constructing an adjacent matrix and a degree matrix based on the similarity matrix after random noise perturbation; obtaining the Laplace matrix based on adjacency matrix and degree matrix; obtaining the first m large eigenvalues and corresponding eigenvectors of Laplace matrices; normalizing the eigenvector to form eigenmatrix; using k-means clustering method to cluster the feature matrix to get the label of clustering. A spectral clustering algorithm is used to calculate the sample similarity between the sample data as the weight value between the data points, and then differential privacy algorithm is used to add random noise of Laplace distribution to the weight value to interfere with the weight value to achieve the purpose of privacy preservation. The interfered data can not only achieve privacy preservation but also ensure the effectiveness of clustering.

The present invention provides an indoor environmental map generation method and device. The method comprises the steps of determining a feasible area in a grid map; segmenting the feasible area by agate segmentation algorithm to obtain a first sub-feasible area and a second sub-feasible area; processing the second sub-feasible area via a spectral clustering algorithm to obtain a plurality of topology areas; fusing a gate area, the first sub-feasible area and the plurality of topology areas and establishing a topology connection relation, thereby obtaining an indoor environmental map. The method is characterized by carrying out the topology segmentation on the grid map, the calculation amount of the topology segmentation is reduced by the gate segmentation algorithm, and by the spectral clustering algorithm, the topology segmentation accuracy is improved. The advantage that the grid map is easy to construct is inherited, the expression of the regional environment categories is realized, at the semantic navigation, the planning efficiency of the paths is high, and the technical problems that an environmental map generation method in the prior art cannot simultaneously satisfy the requirements of easily constructing and realizing the area segmentation are relieved.

The invention provides a space interest point recommendation method of considering both diversity and personalization, and relates to the technical field of space interest point recommendation. The method comprises: constructing a geography-society relationship model; calculating relevance which is of place pairs in the model and on locations and society connection; constructing a relevance matrixW; dividing a user society relationship network graph G constructed in the model; calculating a loss function in dividing; selecting an eigenvector enabling the loss function to be smallest, and dividing vertices in the graph G to obtain k interest point sets with the diversity; and selecting an interest point, which best fits user preference, from each of the k interest point sets to form an interest point recommendation list fusing the diversity and the personalization. According to the space interest point recommendation method of considering both the diversity and the personalization provided by the invention, the geography-society relationship model of interest points, a spectral clustering algorithm and a matrix decomposition algorithm are fused, and the diversity is also consideredwhile interest points recommended for a user are enabled to have a higher accuracy rate.

The invention discloses a method for identifying a cancer molecular subtype based on a spectral clustering algorithm of a sparse similar matrix. The method is characterized in that based on the spectral clustering algorithm of the sparse similar matrix, a cancer molecular subtype prediction model is built by utilizing cancer gene expression profile data as a training set sample; and the prediction model is used for predicting a cancer modular subtype of an independent test set sample, and a cancer sample set is divided into multiple types of molecular subtypes. According to the method, various patients with different prognosis effects are effectively distinguished for high heterogeneity of cancer molecular expression level, and different individual treatment schemes can be made for various cancer patients respectively.

The invention relates to the field of neural network technical methods, in particular to a road network traffic state judgment method based on MFD + spectral clustering + SVM, and the method comprisesthe following specific steps: (1) firstly, carrying out traffic state grade division on a road network MFD by using a spectral clustering algorithm; (2) training an SVM multi-classifier by using thedivided road network MFD parameters, and giving a model classification result precision evaluation method based on a confusion matrix; and (3) finally constructing an Internet of Vehicles simulation platform, selecting a BP neural network classifier as a comparison classifier, and carrying out truth analysis. In the road network traffic state judgment method based on MFD + spectral clustering + SVM, traffic state grade division is performed on the road network MFD by using a spectral clustering algorithm; then an SVM multi-classifier is trained by using the divided road network MFD parameters,a classification result precision evaluation method is given based on a confusion matrix, finally an Internet of Vehicles simulation platform is built, a BP neural network classifier is selected as acomparison classifier, and truth analysis is carried out.

The invention is applicable to the technical field of image processing and provides a semantic mapping method of the local invariant feature of an image. The semantic mapping method comprises the following steps of step A: extracting and describing the local invariant feature of the colorful image; step B: after extracting the local invariant feature, generating a visual dictionary for the local invariant feature extracted from the colorful image on the basis of an algorithm for supervising fuzzy spectral clustering, wherein the visual dictionary comprises the attached relation of visual features and visual words; step C: carrying out semantic mapping and image description on the attached image with the local invariant feature extracted in the step A according to the visual dictionary generated in the step B. The semantic mapping method provided by the invention has the advantages that the problem of semantic gaps can be eliminated, the accuracy of image classification, image search and target recognition is improved and the development of the theory and the method of machine vision can be promoted.

The invention relates to a personalized recommendation method based on dynamic neighboring point spectral clustering. A user-store bipartite network is set up according to user sign-in information; the user-store bipartite network is projected to a user-user one-aside network and a store-store one-aside network, and a node2vec algorithm is used for projecting the two weighted one-aside networks totwo different vector spaces; the spectral clustering algorithm based on dynamic neighboring points is called to cluster user vectors and store vectors obtained above to obtain multiple user clustersand multiple store clusters; the sign-in information existing between single users is converted into a cluster network among the user clusters and the store clusters; a K-means algorithm is used for dividing the one-dimension vector into two classes, the store clusters in the class with the larger sign-in average number are recommended to the user clusters; personalized recommendation is conductedaccording to each user cluster and the recommended store clusters. The accuracy of the recommendation method is improved effectively.

The invention discloses an SAR (Synthetic Aperture Radar) image segmentation method based on sampling learning, mainly solving the problems of huge computation and slow segmentation speed of conventional algorithm. The SAR image segmentation method comprises the following steps: (1) inputting an image to be segmented, and extracting the characteristics; (2) randomly sampling a data set for M times; (3) respectively clustering the data sets of the samples acquired for M times through the spectral clustering algorithm; (4) combining the data of the same type after the clustering implemented for M times, wherein a relative new data set is generated by the combined data of the same type, and training a dictionary for the new data set through KSVD (Singular Value Decomposition) algorithm; (5) calculating sparse codes of the testing samples in the dictionary; (6) calculating the reconstruction error of the testing samples in the dictionary; and (7) determining the labels of the testing samples based on the reconstruction error, so as to obtain the final segmentation result. The SAR image segmentation method based on sampling learning has the advantage of being quick and accurate in segmentation, and can be further applied to target recognition and classification of the SAR image.

The invention discloses an opportunistic network routing method based on spectral clustering community division. The method comprises the following steps of setting the number of network node communities, division classes and the number of nodes and starting carrying out random movement by all nodes; after simulation is carried out for certain period of time, carrying out community division on the nodes in the network through utilization of a spectral clustering algorithm according to contact times and contact duration among the nodes; calculating community reachable probability centrality of a target node, intra-community node reachable probability centrality and a message copy control utility value according to community division results; carrying out movement and encounter by the nodes, and determining whether messages are forwarded or deleted according to the reachable probability centrality and the copy control utility; and periodically updating the community reachable probability centrality, the intra-community node reachable probability centrality and the message copy control utility value. According to the method, through adoption of the spectral clustering algorithm, the community division effect is good, the message transmission is clearly improved, and the network cost is reduced.

The invention discloses a detection method of a mammary image lesion area. The method comprises a step of receiving a mammary image to be detected and preprocessing the mammary image to be detected, astep of carrying out primary cluster segmentation on preprocessed mammary image based on a Nystrom spectral clustering algorithm to obtain a suspicious mammary lesion area, a step of carrying out secondary cluster segmentation on the suspicious mammary lesion area based on a K-means clustering algorithm to obtain a corresponding interested area, and a step of extracting feature information of theinterested area and detecting whether the interested area is a mammary lesion area according to the feature information. The invention also discloses a detection device of a mammary image lesion areaand a computer storage medium. The accuracy of lesion area segmentation in the mammary image can be improved, and thus the accuracy of a detection result of the mammary lesion area is improved.

The invention relates to a software error positioning method and device. The method comprises the steps that through a source code of a testing case driving program, execution path information and a corresponding execution result of the program are acquired, wherein a successful execution path and a failure execution path are involved; an NJW spectral clustering algorithm is adopted for conductingclustering division on the execution path information of the program, an accidental correctness testing case is identified, and the identified accidental correctness testing case is removed; the suspicious degrees of codes are calculated, and a highly-suspicious code of which the suspicious degree is higher than a set suspicious degree threshold value is acquired; an association code associated with the highly-suspicious code is mined on the failure execution path; with the mined association code as a code inspection object, software error positioning is conducted. The software error positioning method has low false report rate and missing report rate on accidental correctness testing case identification, the accuracy rate of software error positioning is increased, and the efficiency ofsoftware error positioning is improved.