936 results about "Singular value decomposition" patented technology

A term-by-document matrix is compiled from a corpus of documents representative of a particular subject matter that represents the frequency of occurrence of each term per document. A weighted term dictionary is created using a global weighting algorithm and then applied to the term-by-document matrix forming a weighted term-by-document matrix. A term vector matrix and a singular value concept matrix are computed by singular value decomposition of the weighted term-document index. The k largest singular concept values are kept and all others are set to zero thereby reducing to the concept dimensions in the term vector matrix and a singular value concept matrix. The reduced term vector matrix, reduced singular value concept matrix and weighted term-document dictionary can be used to project pseudo-document vectors representing documents not appearing in the original document corpus in a representative semantic space. The similarities of those documents can be ascertained from the position of their respective pseudo-document vectors in the representative semantic space.

An audio fingerprinting system and method. A server receives an audio fingerprint of a first audio piece, searches a database for the audio fingerprint, retrieves an audio profile vector associated with the audio fingerprint, updates user preference information based on the audio profile vector, and selects a second audio piece based on the user preference information. The audio fingerprint is generated by creating a matrix based on the frequency measurements of the audio piece, and performing a singular value decomposition of the matrix. To expedite the search of the database and to increase matching accuracy, a subset of candidates in the database is identified based on the most prominent musical notes of the audio piece, and the search is limited to the identified subset. One of the attributes of the audio profile vector is a particular audio class. An identifier for the audio class is generated based on an average of audio fingerprints of the audio pieces belonging to the audio class.

A system and method for transmitting a plurality of input data symbol sub-streams over a plurality of spatial-subspace channels of a sub-carrier between a transmitter and the receiver. The plurality of input data symbol sub-streams are partitioned in a plurality of super-frames of data and weighted by a weight matrix derived from the singular value decomposition of a channel matrix corresponding to the sub-carrier by applying a partial SVD algorithm. The transmitter further inserts subspace training symbols into the plurality of input data symbol sub-streams and the receiver periodically processes the sub-space training symbols during each super-frame of the plurality of super-frames for estimating output data related to the input data symbol stream.

The present invention provides, among other things, automatic identification of near-redundant units in a large TTS voice table, identifying which units are distinctive enough to keep and which units are sufficiently redundant to discard. According to an aspect of the invention, pruning is treated as a clustering problem in a suitable feature space. All instances of a given unit (e.g. word or characters expressed as Unicode strings) are mapped onto the feature space, and cluster units in that space using a suitable similarity measure. Since all units in a given cluster are, by construction, closely related from the point of view of the measure used, they are suitably redundant and can be replaced by a single instance. The disclosed method can detect near-redundancy in TTS units in a completely unsupervised manner, based on an original feature extraction and clustering strategy. Each unit can be processed in parallel, and the algorithm is totally scalable, with a pruning factor determinable by a user through the near-redundancy criterion. In an exemplary implementation, a matrix-style modal analysis via Singular Value Decomposition (SVD) is performed on the matrix of the observed instances for the given word unit, resulting in each row of the matrix associated with a feature vector, which can then be clustered using an appropriate closeness measure. Pruning results by mapping each instance to the centroid of its cluster.

In a technique for video segmentation, classification and summarization based on the singular value decomposition, frames of the input video sequence are represented by vectors composed of concatenated histograms descriptive of the spatial distributions of colors within the video frames. The singular value decomposition maps these vectors into a refined feature space. In the refined feature space produced by the singular value decomposition, the invention uses a metric to measure the amount of information contained in each video shot of the input video sequence. The most static video shot is defined as an information unit, and the content value computed from this shot is used as a threshold to cluster the remaining frames. The clustered frames are displayed using a set of static keyframes or a summary video sequence. The video segmentation technique relies on the distance between the frames in the refined feature space to calculate the similarity between frames in the input video sequence. The input video sequence is segmented based on the values of the calculated similarities. Finally, average video attribute values in each segment are used in classifying the segments.

Text summarizers using relevance measurement technologies and latent semantic analysis techniques provide accurate and useful summarization of the contents of text documents. Generic text summaries may be produced by ranking and extracting sentences from original documents; broad coverage of document content and decreased redundancy may simultaneously be achieved by constructing summaries from sentences that are highly ranked and different from each other. In one embodiment, conventional Information Retrieval (IR) technologies may be applied in a unique way to perform the summarization; relevance measurement, sentence selection, and term elimination may be repeated in successive iterations. In another embodiment, a singular value decomposition technique may be applied to a terms-by-sentences matrix such that all the sentences from the document may be projected into the singular vector space; a text summarizer may then select sentences having the largest index values with the most important singular vectors as part of the text summary.

A method of image processing using kernel regression is provided. An image gradient is estimated from original data that is analyzed for local structures by computing a scaling parameter, a rotation parameter and an elongation parameter using singular value decomposition on local gradients of the estimated gradients locally to provide steering matrices. A steering kernel regression having steering matrices is applied to the original data to provide a reconstructed image and new image gradients. The new gradients are analyzed using singular value decomposition to provide new steering matrices. The steering kernel regression with the new steering matrices is applied to the noisy data to provide a new reconstructed image and further new gradients. The last two steps are repeated up to ten iterations to denoise the original noisy data and improve the local image structure.

The invention relates to an electroencephalogram feature extracting method based on brain function network adjacent matrix decomposition. The current motion image electroencephalogram signal feature extraction algorithm mostly focuses on partially activating the qualitative and quantitative analysis of brain areas, and ignores the interrelation of the bran areas and the overall coordination. In light of a brain function network, and on the basis of complex brain network theory based on atlas analysis, the method comprises the steps of: firstly, establishing the brain function network through a multi-channel motion image electroencephalogram signal, secondly, carrying out singular value decomposition on the network adjacent matrix, thirdly, identifying a group of feature parameters based on the singular value obtained by the decomposition for showing the feature vector of the electroencephalogram signal, and fourthly, inputting the feature vector into a classifier of a supporting vector machine to complete the classification and identification of various motion image tasks. The method has a wide application prospect in the identification of a motion image task in the field of brain-machine interfaces.

The disclosed invention implements SVD-MIMO communication efficiently with a less number of high-load calculation required for singular value decomposition (SVD) processing for a channel matrix. A receiver derives a channel matrix H from a reference signal from a transmitter and acquires downlink transmit weights V and receive weights U^H by SVD of the channel matrix H. The receiver transmits a reference signal weighted with U* to the transmitter, where U* is a conjugate matrix for U as uplink transmit weights. The transmitter receives the reference signal weighted with U* and separates the signal into downlink transmit weights V and a diagonal matrix D, based on unitary matrix properties.

Output-only modal testing of an object. Vibrations are excited in said object and measured by a number of vibration sensitive detectors. From the data of the measurements, a spectral density matrix function is determined. From this density matrix, auto spectral densities for the individual modes are identified performing a decomposition based on the Singular Value Decomposition technique. From the auto spectral densities of the individual modes, natural frequencies and damping ratios for the modes can be estimated, and from the singular vectors of the Singular Value Decomposition, the modes shapes can be estimated.

The invention provides a clustering collaborative filtering recommendation technology based on a singular value decomposition algorithm. The clustering collaborative filtering recommendation technology based on the singular value decomposition algorithm comprises firstly classifying users by using user attributive character values provided by the clustering collaborative filtering recommendation technology based on the singular value decomposition algorithm, and reducing dimension of a user-commodity grade matrix; improving a singular value decomposition (SVD) algorithm which is frequently used in image processing and natural language processing, and using the improved SVD algorithm in a recommendation system; decomposing a grade matrix in a cluster where users are located, and aggregating the decomposed grade matrix so as to fill predicted scores of non-grade items in the grade matrix, calculating similarity of the users in the same cluster by using the filled grade matrix, calculating final predicted scores of a commodity by applying collaborative filtering technologies based on the users and widely applied in the recommendation system, and carrying out final recommendation. The clustering collaborative filtering recommendation technology based on the singular value decomposition algorithm has the advantages of being capable of improving recommendation efficiency of the recommendation system, solving the problems such as data sparsity of the recommendation system, and meanwhile being capable of improving accuracy rate of recommendation of the recommendation system.

Engine response curves (RCs) can be used for streak compensation for printed documents. A feedback control paradigm can be included to effect RC compensation. Singular Value Decomposition (SVD) can be used to represent each RC in the collection of spatial RC data as a linear combination of basis vectors. RCs are approximated by selecting the first few basis vectors, the approximation aiding in noise rejection and reducing computation in the controller by reducing dimensionality of the RC data from gray levels to the number of SVD bases selected. An optimal subset of RCs is selectable from the set of approximated RCs by clustering the SVD weights, the clustered SVD weights producing TRCs that span all engine response RCs generated by a printer. Compensation RCs are constructible using reduced number of bases and clustered SVD weights

A method and system for beamforming communication in a wireless communication system that includes a wireless initiator and a wireless responder is provided. A channel matrix is estimated at the responder. The singular value decomposition of the channel matrix yields the right singular matrix, which is further deconstructed into certain components. The right singular matrix components are quantized in a vector fashion and fed back to the initiator for reconstruction and beamforming communication.

A communication node relays a transmission signal transmitted from a desired source node to a target destination node among multiple source nodes and multiple destination nodes. The communication node includes a first unitary matrix estimation unit that estimates a first unitary matrix by performing singular value decomposition involving one or more channel matrices between the relay node and the source nodes other than the desired source node; a second unitary matrix estimation unit that estimates a second unitary matrix by performing singular value decomposition involving one or more channel matrices between the relay node and the destination nodes other than the target destination node; and a transmission unit configured to transmit a relaying signal generated by multiplying a received signal by the first and second unitary matrices toward the target destination node.

The invention relates to a multi-scale normal feature point cloud registering method. The multi-scale normal feature point cloud registering method is characterized by including the steps that two-visual-angle point clouds, including the target point clouds and the source point clouds, collected by a point cloud obtaining device are read in; the curvature of radius neighborhoods of three scales of points is calculated, and key points are extracted from the target point clouds and the source point clouds according to a target function; the normal vector angular deviation and the curvature of the key points in the radius neighborhoods of the different scales are calculated and serve as feature components, feature descriptors of the key points are formed, and a target point cloud key point feature vector set and a source point cloud key point feature vector set are accordingly obtained; according to the similarity level of the feature descriptors of the key points, the corresponding relations between the target point cloud key points and the source point cloud key points are preliminarily determined; the wrong corresponding relations are eliminated, and the accurate corresponding relations are obtained; the obtained accurate corresponding relations are simplified with the clustering method, and the evenly-distributed corresponding relations are obtained; singular value decomposition is carried out on the final corresponding relations to obtain a rigid body transformation matrix.

The invention discloses an image super-resolution reconstruction method based on multitask KSVD (K singular value decomposition) dictionary learning, mainly aims at solving the problem that the quality of a reconstructed image of the existing method is relatively reduced seriously under a high-magnification factor. The method comprises the following steps of: inputting a training image, filteringthe image to extract characteristics; extracting tectonic characteristics vector sets of small characteristic blocks, and clustering to obtain sample pair sets {(H1, L1), (H2, L2), ..., (HK, LK)} of K to high resolution and low resolution; developing K high-resolution dictionaries Dh1, Dh2, ..., DhK and corresponding low-resolution dictionaries Dl1, Dl2, ..., DlK from the K groups of sample pair sets by means of a KSVD method; encoding low-resolution patterns input in the low-resolution dictionaries Dl1, Dl2, ..., DlK; obtaining an initial reconstruction image by encoding and high-resolution dictionaries Dh1, Dh2, ..., Dh; then implementing local constrained optimization of the initial reconstruction image; and compensating residual errors and implementing global optimization treatment toobtain a final reconstruction image. The image super-resolution reconstruction method based on multitask KSVD dictionary learning has the advantages that the various natural images can be reconstructed, the quality of the reconstructed image can be effectively improved under the condition of a high-magnification factor, and the method can be applied to the recover and identification of human, animal, plant and building and other target objects.

In accordance with the present invention, there is provided a method of determining whether a particular high fidelity vibratory seismic survey phase encoding is likely to be a good one based on an analysis of the eigenvalue structure (i.e., eigenvalues, eigenvalue separation, condition number, and model resolution matrix) of a matrix formed from the Fourier transforms of the sweep signals. Preferably, a singular value decomposition will be used to calculate the eigenvalues. Using this same approach, the condition number and eigenvalues of matrices that are associated with multiple proposed designs can be compared with each other to determine which is likely to be yield the best seismic data. This approach is preferably used either as a component of the advanced planning for a survey or in the field during pre-survey testing. The use of the instant approach to determine an optimal phase encoding scheme is also taught.

A forecasting apparatus for predicting future events includes a forecast processing data configuring section for configuring a data matrix including previously accumulated historical data and unknown forecast data, the data matrix having the unknown forecast data as missing elements, and a forecast processing section for estimating values of the missing elements representative of the unknown forecast data by performing singular value decomposition of the data matrix configured by the forecast processing data configuring section.

The invention discloses an image classification method based on hierarchical SIFT (scale-invariant feature transform) features and sparse coding. The method includes the implementation steps: (1) extracting 512-dimension scale unchanged SIFT features from each image in a data set according to 8-pixel step length and 32X32 pixel blocks; (2) applying a space maximization pool method to the SIFT features of each image block so that a 168-dimension vector y is obtained; (3) selecting several blocks from all 32X32 image blocks in the data set randomly and training a dictionary D by the aid of a K-singular value decomposition method; (4) as for the vectors y of all blocks in each image, performing sparse representation for the dictionary D; (5) applying the method in the step (2) for all sparse representations of each image so that feature representations of the whole image are obtained; and (6) inputting the feature representations of the images into a linear SVM (support vector machine) classifier so that classification results of the images are obtained. The image classification method has the advantages of capabilities of capturing local image structured information and removing image low-level feature redundancy and can be used for target identification.

A hydroelectric generating set vibration signal noise reduction method based on mode autocorrelation analysis of SVD and VMD comprises the steps of constructing a Hankel matrix of a set vibration signal and performing singular value decomposition (SVD), selecting an effective singular value based on a mean value filtering strategy for reconstructing the signal, and realizing pre-filtering; performing decomposition through variational mode decomposition (VMD) for obtaining a series of mode functions, calculating an autocorrelation function of each mode component, selecting effective mode components according to an energy set of the autocorrelation function, and obtaining a signal after noise reduction through adding all effective mode components. According to the hydroelectric generating set vibration signal noise reduction method provided by the invention, a noise reduction experiment is carried out through simulation analysis and actual measurement of a vibration signal; and a result represents a fact that the method has relatively high noise reduction performance and can effectively improve hydroelectric generating set vibration signal analysis precision.

The invention relates to a method for extracting and fusing time, frequency and space domain multi-parameter electroencephalogram characters, which comprises the following steps: 1) collecting an electroencephalogram signal; 2) performing data pre-processing on the electroencephalogram signal; 3) extracting Kc complexity, approximate entropy and wavelet entropy from the pre-processed data; 4) on the basis of AMUSE algorithm, acquiring an electroencephalogram singular value decomposition matrix parameter; 5) performing character selection on the time, frequency and space domain character parameters for the extracted Kc complexity, approximate entropy, wavelet entropy and electroencephalogram singular value decomposition matrix parameters; 6) utilizing a SVM classifier to fuse and classify the four parameters of the time, frequency and space domains after the character selection. According to the method provided by the invention, the Kc complexity, the approximate entropy, the wavelet entropy and the electroencephalogram singular value decomposition matrix parameter can be selected for comprehensively presenting electroencephalogram character information, and then subsequent effective fusion is performed, so that effective support and help can be supplied to early diagnosis assessment for the brain functional disordered diseases, such as, Alzheimer disease, mild cognitive impairment, and the like.

The invention discloses a low-rank structure-based sparse compressive sensing MRI (Magnetic Resonance Imaging) image reconstruction method, which mainly solves the problem of difficulty in accurate recovery of an MRI image existing in the conventional technology. The method comprises the following implementation steps: initially recovering the MRI image by using the conventional compressive sensing and looking for a similar block matrix from the image to form an index set; performing singular value decomposition on the similar block matrix and calculating a threshold, and performing threshold calculation on a singular value by using the threshold to obtain an after-threshold singular value; and optimizing the MRI image by using the after-threshold singular value, i.e., circularly performing processes of updating of the similar block matrix and the index thereof, similar block matrix singular value decomposition and threshold and singular value threshold calculation on the MRI image to obtain a final recovery image. The recovered MRI image is clearer and has sharper edges; and the method can be used for processing a medical image.

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 non-convex low-rank reconstruction method for rapid MR imaging. An MR image data reconstruction mathematic model based on low-rank prior information of non-local similar image blocks is established, and iterative solution is carried out on the model in a direction alternative iteration method; a non-convex p norm of the low-rank matrix of the non-local image model with the low-rank prior information is solved by deposition and iteration of Taylor first-order approximation and the singular value, a similar image block is obtained, and a reconstruction image is solved via iteration by increasing the auxiliary variable and separating the variable. The image prior information is used to combine the non-local similarity with the low-rank characteristic of the image block, the Fourier transform and the characteristic of the low-rank matrix are used to simplify the calculation process, the complexity of algorithm is reduced, the performance of the reconstructed MRI images by part of K space data is improved, the image can be reconstructed more accurately with less scanning and measurement, pseudo shadows of the images are reduced, and rapid MRI is realized.

An apparatus and method for controlling amount of information to be fed back in a multiple antenna system in a multi-user environment. The apparatus includes a channel estimator which estimates channel values by using an input signal; a Singular Value Decomposition (SVD) operator which decomposes singular values from the estimated channel values; and a feedback determining unit which uses the singular values to compute channel capacity that can be obtained according to the amount of channel information to be fed back, and selects the channel information to be fed back by using the channel capacity.

Channel state information in a closed-loop, multiple-input, multiple-output wireless networks is fed back from each mobile station to a base station by first determining a transmit covariance matrix R, and applying a singular value decomposition (SVD) R=UΣV^H, where U, V are left and right singular vector matrices, Σ is a diagonal matrix with singular values. The matrix V includes column vectors V. A beamforming vector v_max=[1 exp(jΦ)exp(j2Φ) . . . exp(jΦ)]/√{square root over (N)}] is approximated by the column vector V having a maximum magnitude, where Φ is a real number. Then, only the angle Φ is fed back using a phase modulation mapping of the components exp(jΦ) onto the associated subcarrier.

The invention discloses a three-dimensional position reconstructing method based on an ISAR (inverse synthetic aperture radar) image sequence for a scattering point, which comprises the following four links: a target ISAR image gives the distribution information of a target strong scattering point in a radial direction and a transverse direction based on a distance-Doppler high resolution basic principle; data correlation gives a corresponding relationship between two-dimensional projection points in the ISAR image sequence and is realized by utilizing a flight path initialization method through extracting the one-dimensional radial distance information of all the scattering points in an image sequence; an observing matrix is obtained through the following steps: obtaining a target ISAR image sequence through a period of time of sampling, and after the data correlation, and combining the two-dimensional position coordinates of all the corresponding projection points in a sequence to form the observing matrix, so as to form three-dimensional reconstructed known information; and a position matrix is solved through the following step: solving the optimal estimation of a three-dimensional position matrix of a target scattering point from a subspace through carrying out singular value decomposition on the observing matrix and by utilizing a rank theory and using the orthogonality of a projection space as a constraint condition, so as to obtain a target three-dimensional reconstructed image.

A system and method for rendering a virtual sound source using a plurality of speakers in an arbitrary arrangement includes expanding a wave field of the virtual sound source and a composite wave field generated by the plurality of speakers into multi-pole expansions, and equating the multi-pole expansions to each other to form a continuous equation. A set of constraints is applied to the continuous equation that minimize an error in a sweet spot region resulting in a linear system of equations. The linear system of equations is solved using singular-value decomposition to arrive at a pseudo-inverse representing a multi-dimensional transfer function between the virtual sound source and the plurality of speakers. The multi-dimensional transfer function is then applied to a signal of the virtual sound source to render the virtual sound source in the sweet spot region using the plurality of speakers in the arbitrary arrangement.

The invention discloses an image super-resolution reconstruction method based on self-similarity and structural information constraint. The image super-resolution reconstruction method based on the self-similarity and the structural information constraint comprises the achieving steps: (1) taking z images from an image base, carrying out imitating quality degradation on each image, generating a low-resolution image, and constructing a dictionary training sample set; (2) in the dictionary training sample set, learning a pair of high resolution ratio dictionary and low resolution ratio dictionary through a kernel singular value decomposition (K-SVD) method; (3) for a to-be-processed low-resolution image Xt, with scale rotation transform utilized, searching k similar blocks {p1,p2,...,pk} which are mostly similar with an image block xi; (4) carrying out constraint solution on the image block xi through the obtained k similar blocks to obtain a sparse presentation coefficient A; (5) obtaining k reconstruction results through the sparse presentation coefficient A combined with a high-resolution dictionary DH; (6) utilizing a low rank presentation model, amending a similarity degree of the reconstruction results with the similar blocks {p1,p2,...,pk} under the low resolution utilized; (7) obtaining a final result through the amended similarity degree combined with the reconstruction results; and repeating the steps in sequence and obtaining a final high-resolution image YH. The image super-resolution reconstruction method based on the self-similarity and the structural information constraint has the advantages that structural information of the reconstruction results keeps good, and the image super-resolution reconstruction method can be used for image recognition and target classification.

The present invention is to provide a wireless communication system and a receiving device each capable of restraining a circuit scale and obtaining an excellent error rate performance. In a wireless communication system in which a transmitting device and a receiving device communicate in an orthogonal frequency division multiplexing system, the transmitting device includes a partition unit partitioning a transmission data sequence into a plurality of data streams and a mapping unit mapping the plurality of data streams to each of transmitting antennas by using, for a precoding matrix, columns, corresponding to a stream count of the plurality of data streams, of a right singular matrix acquired by singular value decomposition of a channel matrix, and the receiving device includes a decoding unit that Viterbi-decodes the received signals by weighting a path metric by using a weighting coefficient corresponding to a signal-to-noise ratio obtained from the present channel matrix.