4520 results about "Dimensionality reduction" patented technology

System and method for dimension reduction and for channel and interference condition feedback in a Multi-User Multiple-Input-Multiple-Output (MU MIMO) wireless communication systems. The method for dimension reduction includes determining a number of virtual antennas, v_k, for beamformees pertaining to a multi user (MU) transmission group, reducing the dimension of said matrix by selecting a sub-set of antennas or by Eigen mode selection and sending a reduced dimension candidate transmit beamforming matrix or an effective channel matrix to a beamformer. The method for channel and interference condition feedback Interference condition includes sending to a beamformee metric indicative of the interference level from other streams.

A depolarization waveform classifier based on the Modified lifting line wavelet Transform is described. Overcomes problems in existing rate-based event classifiers. A task for pacemaker/defibrillators is the accurate identification of rhythm categories so correct electrotherapy can be administered. Because some rhythms cause rapid dangerous drop in cardiac output, it's desirable to categorize depolarization waveforms on a beat-to-beat basis to accomplish rhythm classification as rapidly as possible. Although rate based methods of event categorization have served well in implanted devices, these methods suffer in sensitivity and specificity when atrial/ventricular rates are similar. Human experts differentiate rhythms by morphological features of strip chart electrocardiograms. The wavelet transform approximates human expert analysis function because it correlates distinct morphological features at multiple scales. The accuracy of implanted rhythm determination can then be improved by using human-appreciable time domain features enhanced by time scale decomposition of depolarization waveforms.

The invention discloses a power grid security situation predicting method based on an improved deep learning model and belongs to the technical field of power system safety. The power grid security situation predicting method includes: performing power grid security situation evaluation through power grid data collection and preprocessing; aiming at the characteristic that indicator data of power grid security situation evaluation are high in relevance and dimension, providing an improved self-coding network method to lower dimension of the indicator data, and utilizing a data sample after dimension reduction and a power grid security situation value corresponding to a next time monitoring point; adopting an improved deep belief network to build a deep learning situation predicting model with multi-input and multi-output for power grid security situation prediction. By the power grid security situation predicting method, speed and accuracy of power grid security situation prediction can be improved effectively.

The invention discloses a block chain metadata storage system, and a storage method and a retrieval method thereof. The system comprises at least one main node and a plurality of secondary nodes which are respectively connected with the Internet and construct a distributed shared network via a block chain. The main node a data receiving module used for receiving source data from the outside; a distributed storage module connected with the data receiving module, and used for storing the source data in a distributed manner; a dimensionality reduction and preservation module connected with the distributed storage module, and used for performing preservation operation on metadata to form a data fingerprint after the metadata is formed by performing dimensionality reduction on the source data; and a data storage module connected with the dimensionality reduction and preservation module, and used for writing the data fingerprint into the block chain and performing whole network issue by using a consensus process of the block chain. Each secondary node comprises a data consensus module used for receiving and storing the data fingerprint issued by the block chain and completing the consensus process of the block chain.

The invention discloses a short text clustering method based on deep semantic feature learning. The method includes the steps that dimensionality reduction representation is performed on original features under the restraint of local information preservation through traditional feature dimensionality reduction, binarization is performed on an obtained low-dimension actual value vector, and error back propagation is performed with the binarized vector being supervisory information of a convolutional neural network structure to train a model; non-supervision training is performed on a term vector through an outer large-scale corpus, vectorization representation is performed on all words in text according to the word order, and the vectorized words serve as implicit semantic features of initial input feature learning text of the convolutional neural network structure; after deep semantic feature representation is obtained, a traditional K-means algorithm is adopted for performing clustering on the text. By means of the method, extra natural language processing and other specialized knowledge are not needed, design is easy, deep semantic features can be learnt, besides, the learnt semantic features have unbiasedness, and good clustering performance can be achieved more effectively.

The invention discloses a network attack abnormality detection method, and relates to the technical field of information security. The technical point of the network attack abnormality detection method disclosed by the invention is as follows: a step 1, deploying a traffic data collection device at a network traffic aggregation node; a step 2, extracting a network behavior feature value from the collected traffic data; a step 3, carrying out dimensionality reduction and standardization on the network behavior feature value; a step 4, determining normal network behavior feature values, and establishing a normal behavior model based on a set of the normal network behavior feature values; and a step 5, detecting other network behavior feature values based on the normal behavior model, and judging whether abnormal network behaviors occur; and when detecting abnormal network behaviors, updating the normal behavior model according to a new normal network behavior feature value.

The invention provides a universal circuit breaker mechanical fault diagnosis method based on feature fusion of vibration and sound signals. The method includes steps of 1, collecting machine vibration signals and machine sound signals during an engaging and disengaging process of a universal circuit breaker; 2, adopting an improved wavelet packet threshold value denoising algorithm for denoising; 3, adopting a complementary total average empirical mode decomposition algorithm for extracting a plurality of solid mode function components reflecting state information of engagement and disengagement actions of the circuit breaker from the denoising signals; 4, determining the number Z of the solid mode function components; 5, calculating the energy ratio, the sample ratio and the power spectrum entropy as three types of features; 6, adopting a combination core principal component analysis method for performing dimension reduction on a feature sample with unified three types of features of the vibration and the sound signals and obtaining M principle components; 7, establishing a related vector machine based sequence binary tree multiple classifier model.

The invention relates to the technical field of digital image processing and pattern recognition, discloses a fuzzy clustering steel plate surface defect detection method based on pre classification and aims to overcome defects of judgment missing and mistaken judgment by the existing steel plate surface detection method and improve the accuracy of steel plate surface defect online real-time detection effectively during steel plate surface defect detection. The method includes the steps of 1, acquiring steel plate surface defect images; 2 performing pre classification on the images acquired through step 1, and determining the threshold intervals of image classification; 3, classifying images of the threshold intervals of the step 2, and generating white highlighted defect targets; 4, extracting geometry, gray level, projection and texture characteristics of defect images, determining input vectors supporting a vector machine classifier through characteristic dimensionality reduction, calculating the clustering centers of various samples by the fuzzy clustering algorithm, and adopting the distances of two cluster centers as scales supporting the vector machine classifier to classify; 5, determining classification, and acquiring the defect detection results.

The invention relates to a voice-and-facial-expression-based identification method for dual-modal emotion fusion. The method comprises: S1, audio data and video data of a to-be-identified object are obtained; S2, a face expression image is extracted from the video data and segmentation of an eye region, a nose region, and a mouth region is carried out; S3, a facial expression feature in each regional image is extracted from images of the three regions; S4, PCA analysis and dimensionality reduction is carried out on voice emotion features and the facial expression features; and S5, naive Bayesian emotion voice classification is carried out on samples of two kinds of modes and decision fusion is carried out on a conditional probability to obtain a final emotion identification result. According to the invention, fusion of the voice emotion features and the facial expression features is carried out by using a decision fusion method, so that accurate data can be provided for corresponding conditional probability calculation carried out at the next step; and an emotion state of a detected object can be obtained precisely by using the method, so that accuracy and reliability of emotion identification can be improved.

Disclosed is a copper sheet and strip surface defect detection method based an on-line sequential extreme learning machine. The method includes the following steps that a copper sheet and strip surface image is captured through an image capturing module; the captured copper sheet and strip surface image is enhanced according to the median filtering method with the masking size of 7*7 to reduce noise in the copper sheet and strip surface image and the effect of the noise on the quality of the surface image; the copper sheet and strip surface image is subject to tophat transform treatment to reduce the effect of uneven illumination; a copper sheet and strip surface image pre-detection method based on eight-neighborhood difference values is adopted; defects in the surface image are segmented according to an image segmentation method, wherein it is judged that the copper sheet and strip surface image has the surface defects after pre-detection; geometrical characteristics, gray characteristics, shape characteristics, texture characteristics and other characteristics of each defect are extracted, and copper sheet and strip surface defect characteristic dimensions are subject to optimization and dimensionality reduction according to the principal component analysis method; a copper sheet and strip surface defect classifier based on the on-line sequential extreme learning machine is designed, and samples are used for training; characteristics of the copper sheet and strip surface image to be detected are extracted to identify types of the surface defects.

The invention provides a bearing fault classification diagnosis method based on sparse representation and an LDM (large margin distribution machine), overcomes the defects that signal decomposition is incomplete, a reconstructed signal cannot better keep features of an observed signal and the like in the conventional single-channel mechanical compound fault diagnosis method. According to the method, signal conversion from one dimension to high dimension is realized with a CEEMD (complete ensemble empirical mode decomposition) method, the decomposition completeness is guaranteed, and a mode mixing phenomenon is inhibited; meanwhile, a dimensionality reduction method based on sparse representation is introduced into a feature extracting and processing process of a blind source signal, data are subjected to sparse reconstruction through sparse representation, and data feature information is extracted from global data, so that the reconstructed signal can better keep the data features of the observed signal; further, the LDM classification method is introduced into a model fault type classification processing process of a to-be-detected bearing, and the accuracy and effectiveness of bearing fault diagnosis can be improved by aid of the generalization ability of the LDM classification method.

The invention discloses a sparse dimension reduction-based spectral hash indexing method, which comprises the following steps: 1) extracting image low-level features of an original image by using an SIFT method; 2) clustering the image low-level features by using a K-means method, and using each cluster center as a sight word; 3) reducing the dimensions of the vectors the sight words by using a sparse component analysis method directly and making the vectors sparse; 4) resolving an Euclidean-to-Hamming space mapping function by using the characteristic equation and characteristic roots of a weighted Laplace-Beltrami operator so as to obtain a low-dimension Hamming space vector; and 5) for an image to be searched, the Hamming distance between the image to be searched and the original image in the low-dimensional Hamming space and using the Hamming distance as the image similarity computation result. In the invention, the sparse dimension reduction mode instead of a spectral has principle component analysis dimension reduction mode is adopted, so the interpretability of the result is improved; and the searching problem of the Euclidean space is mapped into the Hamming space, and the search efficiency is improved.

The invention discloses a method for estimating target arrival angle of a multiple input multiple output (MIMO) radar, which mainly solves the problem of large signal processing capacity in the target positioning process of MIMI radar. The method comprises the following steps of: 1) obtaining a virtual array of echo of each receiving antenna by a matched filter bank; 2) constructing a data conversion matrix and a dimension reduction array according to position of transmitting and receiving array; (3) reducing dimension of the virtual array data by the dimension reduction array to obtain an effective array after dimension reduction; 4) constructing two sub-arrays by rotational variance of effective array, and deriving covariance matrix of data; 5) decomposing eigenvalue of covariance matrix to obtain signal sub-spaces corresponding to two sub-arrays; 6) deriving rotational invariant relationship matrix by least square method to obtain arrival angle of target. The dimension reduction matrix form constructed by the method has versatility; the computation quantity is reduced by the dimension reduction of data and the ESPRIT (estimating signal parameter via rotational invariance techniques) algorithm; the computation speed of MIMO radar is increased; and the real-time signal processing of the MIMO radar is made easier.

The invention discloses an improved high-resolution remote sensing image classification method based on deep learning. On the basis of the deep learning theory, a seven-layer convolutional neural network is designed; a high-resolution remote sensing image sample is inputted into the network to carry out network training and last two full connection layers obtained by learning are outputted as twodifferent high-level features of the remote sensing image; dimension reduction is carried out by using a principal component analysis for the output of the fifth pooling layer of the network, whereinthe result after dimension reduction is used as a third high-level feature of the remote sensing image; the three kinds of high-level features are fused in series; and then an effective logistic-regression-based classifier is designed to classify the remote sensing image. According to the invention, feature extraction is carried out on the high-resolution remote sensing image based on the deep learning theory and the features obtained by learning have high expressive force and robustness. Besides, the extracted high-level features are fused and the fused feature is inputted into the logistic regression classifier, so that the good classification result is obtained.

The invention discloses a short period load prediction method based on kernel principle component analysis and a random forest. The a short period load prediction method comprises the following steps of: (1) analyzing and selecting data influencing load prediction precision of a day to be predicted in an operational electric power system, and preliminarily constructing training and prediction sample sets; (2) utilizing kernel principle component analysis to carry out dimensionality reduction on training sample data; (3) utilizing a random forest model to train the training sample data after the dimensionality reduction, and obtaining the random forest model after the training; and (4) inputting prediction sample data into the random forest model after the training, and carrying out short period load prediction of the day to be predicted. The short period load prediction method has the advantages that the kernel principle component analysis and the random forest model are combined for carrying out short period load prediction on the electric power system, the prediction precision, efficiency and data rationality are improved.

The invention discloses a rolling bearing fault diagnosis method based on composite multi-scale permutation entropies, and belongs to the technical field of fault diagnosis. The method comprises the following steps: measuring a vibration signal of a faulty object; extracting composite multi-scale permutation entropies from the vibration signal; reducing the dimension of the composite multi-scale permutation entropies with use of a Laplacian score; taking the first multiple composite multi-scale permutation entropies with low scores after dimension reduction as fault feature vectors and dividing the fault feature vectors into multiple training samples and multiple test samples; inputting the multiple training samples into a multi-fault classifier established based on a support vector machine to perform learning so as to classify the test samples; and identifying the working mode and the fault type of the faulty object according to the classifying result. According to the fault diagnosis method disclosed by the invention, feature extraction is highly innovative, and the degree of identification is high in the process of fault mode identification.

The invention discloses the technical field of wind power prediction, particularly, relates to an ultrashort-term power prediction method. The method comprises the following steps of: firstly, acquiring the wind speed, the wind direction and the wind power of a wind power farm to form a sample set; then, preprocessing the data of the sample set; reducing dimensions of the preprocessed sample set by a depth autocoder network; and finally, training a relevance vector machine regression model by the sample set with reduced dimensions, and predicting the ultrashort-term wind power through the trained relevance vector machine regression model. The method reduces the training time of a prediction model, satisfies the requirements on precision and real-time property in system status estimation, and enables the prediction method to be more accurate.

The present invention discloses a multi-year regulating storage reservoir optimal scheduling method considering runoff uncertainty. The method comprises the following steps of: (1) analyzing historical reservoir runoff data; (2) establishing a runoff random simulation model, and generating a reservoir simulation annual runoff sequence; (3) simulating annual inflow runoff sequence classification and performing intra-annual distribution to obtain a three-year inflow runoff process sequence on a month scale; (4) establishing a multi-year regulating storage reservoir optimal scheduling model that considers multiple water demands; and (5) reducing dimension by using an optimization algorithm and optimizing a resolution model so as to obtain an optimal discharging and water supply decision. According to the method provided by the present invention, the scheduling with maximized comprehensive benefits of long-term multi-target operation can be performed on a multi-year regulating storage reservoir by considering runoff uncertainty, so that the method is suitable for promotion in long-term optimal scheduling of the multi-year regulating storage reservoir in China.

The invention discloses a power consumption data anomaly detection model based on an isolated forest algorithm. The model comprises a feature extraction module, a feature dimension reduction module, an isolated forest calculation module, an expert sample module and a secondary training module, wherein the feature extraction module extracts the time series of the user's power consumption data fromthe original data set as an initial feature set, and then carries out dimensionless and feature selection processing on the initial feature set; the feature dimension reduction module adopts principalcomponent analysis and self-coding network method to reduce the dimension of the initial feature set to get the effective feature set; the isolated forest computing module uses isolated forest algorithm to calculate the outlier score of each user to determine whether the user data is abnormal or not. The electric power data anomaly detection model based on the isolated forest algorithm of the invention is an unsupervised electric power data anomaly detection model, which not only can quickly process a large amount of data, but also can adapt to the situation of lack of training samples, and can better meet the practical requirements of the electric power department.

The invention belongs to the field of computer vision and pattern recognition and discloses a moving target tracking method based on an improved multi-example learning algorithm. Firstly, a random measurement matrix is designed according to the compression perception theory. Then a multi-example learning algorithm is used to sample an example in a current tracking result small neighborhood to form a positive package, and at the same time, sampling an example is carried out in a large neighborhood ring to obtain a negative package. For each example, the characteristic of a character target is extracted at an image surface, and the random measurement matrix is utilized to carry out dimensionality reduction on the characteristic. According to the extracted example characteristic, online learning weak classifiers are utilized, and weak classifiers with strong discrimination ability are selected from a weak classification pool to form a strong classifier. Finally, when a new target position is tracked, according to a similarity score of the current tracking result and a target template, the online adaptive adjustment of classifier update degree parameters is carried out. According to the method, a problem that a tracking result in the existing algorithm is easily affected by an illumination change, an attitude change, the interference of a complex background, target fast motion and the like is solved.

The invention discloses a wearable artery detection device comprising an electrocardiogram sensor module, a heart sound sensor module and a pulse wave sensor module all carrying independent relative position sensors inside, as well as a health management module. The health management module is used for giving composite learning indexes. An electrocardiogram sensor is provided with at least three leads located in the lest chest cavity and is packaged in a small patch together with a heart sound sensor, and a reference position sensor is arranged in the small patch to serve as reference coordinates of the relative position sensors. A data processing method of the wearable artery detection device includes: dividing all dynamic physiological parameters obtained by calculation into multiple time windows according to rhythm of dynamic electrocardiograms, inputting high-dimensional feature vectors of the time windows into a deep learning network, and performing supervised deep learning to obtain the composite learning indexes subjected to dimension reducing processing for performing learning processing at the next step prior to obtaining final learning results.

The invention discloses a cost optimization unmanned aerial vehicle base station deployment method based on an improved genetic algorithm, and mainly solves the problem that the unmanned aerial vehicle base station deployment cost is difficult to optimize in the prior art. The realization method comprises the following steps: 1) establishing a ground wireless communication coverage model of the unmanned aerial vehicle base station; 2) calculating the maximum coverage radius and the optimal hovering height of the unmanned aerial vehicle base station in the unmanned aerial vehicle base station ground wireless communication coverage model scene; 3) deploying the unmanned aerial vehicle base stations at the optimal hovering height, enabling the deployment problem to be reduced from three-dimensional dimensionality to a two-dimensional plane, establishing an unmanned aerial vehicle base station deployment optimization model taking unmanned aerial vehicle base station deployment number optimization as a target, and solving the model to obtain an optimal chromosome; 4) converting the optimal chromosome into a corresponding unmanned aerial vehicle base station coordinate set to obtain an optimal unmanned aerial vehicle base station deployment scheme, reducing the complexity of the deployment problem, improving the solution accuracy, and being applicable to communication network deployment planning, temporary communication network construction and disaster area emergency communication.

The invention discloses an image search method and apparatus. The image search method comprises the steps of obtaining a target region of interest of a to-be-searched image; extracting a local eigenvector and a deep learning eigenvector of the target region of interest; executing feature dimension reduction processing on the local eigenvector and the deep learning eigenvector by correspondingly utilizing a preset locally-weighted index and a preset depth-weighted index respectively, and performing feature fusion on the local eigenvector subjected to the dimension reduction and the deep learning eigenvector subjected to the dimension reduction by utilizing a preset splicing weighted index, thereby obtaining a target eigenvector which improves the feature description precision of the target region of interest; and performing a search according to the target eigenvector, thereby obtaining a search result based on the to-be-searched image.