289 results about "Matrix representation" patented technology

An apparatus includes a first section operable to detect polarization-sensitive radiation emitted by an object, and a second section operable to determine a Jones matrix based on information obtained by the first section from the polarization-sensitive radiation. The second section thereafter transforms the Jones matrix into a Mueller matrix, the Mueller matrix being representative of properties of the object.

A resource allocation method for use in a wireless ATM network comprises receiving on a wireless signaling channel a request for access to a shared frequency-time sliced wireless medium. A channel matrix is then searched for a set of available frequency-time slots. The channel matrix represents a time frame within the shared frequency-time sliced wireless medium, and is used to keep track of resource allocation in the time-frequency sliced medium. The set of available time-slots is then allocated if the allocation does not violate a frequency switching constraint, and if the set of available frequency-time slots contains a number of slots no smaller than a requested number of slots. In a preferred embodiment of the invention, the searching step uses a greedy resource allocation strategy to search a channel-chunk matrix comprising a list of contiguous chunks of available time slots in each frequency of the shared frequency-time sliced wireless medium. The greedy resource allocation strategy comprises the following successive allocation steps: searching for a single contiguous set of available time slots in a single frequency, where the size of the set of available slots is equal to the requested size [60]; searching for a single contiguous set of available time slots in a single frequency, where the size of the set of available slots is greater than the requested size [64]; searching for separate chunks of available time slots in a single frequency [66]; and searching for separate chunks of available time slots in multiple frequencies [68]. Each allocation step comprises checking whether the allocation violates a frequency switching constraint.

A method and apparatus for representing complex data schemas and generating type validated output documents in a markup language. The methods apply to transforming document type definitions into extensible markup language coded information that can readily accommodate logical constraints imposed by recursion or repetition within the DTD structure. Furthermore, non-determinism arising from repetition or recursion in a data schema is resolved by traversal path coding using a matrix representation of the data schema.

An extension of an inverse inference search engine is disclosed which provides cross language document retrieval, in which the information matrix used as input to the inverse inference engine is organized into rows of blocks corresponding to languages within a predetermined set of natural languages. The information matrix is further organized into two column-wise partitions. The first partition consists of blocks of entries representing fully translated documents, while the second partition is a matrix of blocks of entries representing documents for which translations are not available in all of the predetermined languages. Further in the second partition, entries in blocks outside the main diagonal of blocks are zero. Another disclosed extension to the inverse inference retrieval document retrieval system supports automatic, knowledge based training. This approach applies the idea of using a training set to the problem of searching databases where information that is diluted or not reliable enough to allow the creation of robust semantic links. To address this situation, the disclosed system loads the left-hand partition of the input matrix for the inverse inference engine with information from reliable sources.

A method and apparatus for decoding two-dimensional barcode symbols from a data stream generated by scanning a two-dimensional barcode with a one-dimensional scanner. A decoding process according to one embodiment of the invention involves coupling the one-dimensional scanner to a serial or other port of a host computer in order to provide a data stream to a software decoder resident in the host computer. The software decoder performs a series of iterative steps in which the data stream is sampled at regular intervals to convert the data stream to a series of elements (e.g., bars and spaces). Barcode symbol data characters are then derived from the elements. A symbol matrix is next constructed, the matrix being representative of the two-dimensional barcode symbol.

System and methods are disclosed for providing answers to search queries, and for searching using association data without requiring keyword matching. Datasets representing objects and their properties are created from unstructured data sources based on natural language analysis methods, and can be used to answer queries about objects or properties of objects. Implementations include general information search engines and embodiments for searching products, services, people, or other objects without knowing the names of such objects, or searching for information about known objects by using either keyword-based queries or natural language queries such as asking questions. System and methods are also provided for creating a structured or semi-structured representation of various unstructured data, in contrast to the conventional term-vector or term-document matrix representation.

A delay measurement technique according to an embodiment according to the present invention is based on the precept, ascertained by the inventors, that a link between network nodes will often contribute to the delay encountered between several different pairs of network nodes. Such a technique identifies the path between each pair of nodes by a list of links that form the path. Paths that are orthogonal are treated as being necessary for describing the delays encountered between nodes, and, once the requisite set of orthogonal paths has been derived, all other paths can be described in terms of one or more of these orthogonal paths. Such a technique also lends itself to matrix representation of the paths, and the use of matrix manipulation techniques in deriving delay and jitter.

A method for tone mapping high dynamic range images for display on low dynamic range displays wherein high dynamic range images are first accessed. The high dynamic range images are divided the into different regions such that each region is represented by a matrix, where each element of the matrix is a weight or probability of a pixel value. An exposure of each region is determined or calculated by estimating an anchor point in each region such that most pixels in each region are mapped to mid grey and the anchor points are adjusted to a key of the images to preserve overall brightness. The regions are then placed or mapped to zones and exposure values are applied to the regions responsive to the weight or probability. The regions are fused together to obtain a final tone mapped image.

A computer implemented method and apparatus for storing and retrieving data embedded into the surface of a 3D printed object is described. The method and apparatus develops an electronic file used for printing a 3D object which embeds as structure into the 3D object, a 3D symbol matrix representative of data to be printed concurrently with the 3D object, such as a 3D barcode. A selected symbology is used for making the symbol matrix in accordance with the type of printing process to be used to print the 3D object.

A stream of ordered information, such as, for example, audio, video and / or text data, can be windowed and parameterized. A similarity between the parameterized and windowed stream of ordered information can be determined, and a probabilistic decomposition or probabilistic matrix factorization, such as non-negative matrix factorization, can be applied to the similarity matrix. The component matrices resulting from the decomposition indicate major components or segments of the ordered information. Excerpts can then be extracted from the stream of ordered information based on the component matrices to generate a summary of the stream of ordered information.

A method, apparatus, computer product and structure is presented for representing and managing large amounts of information concerning networks of elements. While being useful for communication networks, it can be also usefully deployed in the context of other networks such as distribution and transportation networks. The method uses a hierarchical construct called “catalog”—a set of elements (which could be “atomic” elements or catalogs themselves)—to organize information about physical or abstract entities relevant for modeling the network. A matrix construct whose rows and columns constitute such elements are used to model connections at different levels of abstraction. A common framework and representation provided using these two constructs is shown to be useful for visualization, administration, configuration, modeling, monitoring and manipulation of the network.

A delay measurement technique according to an embodiment according to the present invention is based on the precept, ascertained by the inventors, that a link between network nodes will often contribute to the delay encountered between several different pairs of network nodes. Such a technique identifies the path between each pair of nodes by a list of links that form the path. Paths that are orthogonal are treated as being necessary for describing the delays encountered between nodes, and, once the requisite set of orthogonal paths has been derived, all other paths can be described in terms of one or more of these orthogonal paths. Such a technique also lends itself to matrix representation of the paths, and the use of matrix manipulation techniques in deriving delay and jitter.

The invention discloses a method and a device for adjusting an artificial neural network (ANN). The ANN comprises multiple neurons, and the connection relationship among the neurons is presented through a connection weight matrix. The method comprises a pruning step in which n unimportant weights in all N weights of a first trained connection weight matrix are set to be zero, a step of retraining without a mask in which a second connection weight matrix through pruning is retrained in a condition in which any weight is not forced to be constrained to be zero, a step of mask generation in which a matrix-shaped mask is generated according to a third connection weight matrix through the retraining without the mask and a step of retraining with the mask in which the mask matrix is used for retraining the third connection weight matrix. Thus, the mask is dynamically adjusted through adding the step of generating the mask via the matrix without mask retraining in a retraining stage, error pruning in the pruning process is corrected and restored, and the performance of the compressed neural network is enhanced.

The invention discloses an enterprise entity relation extraction method based on a convolutional neural network. The method comprises the steps of a relation corpus building stage, wherein an initial seed relation pair set is built artificially, and by means of an internet search engine and a Bootstrapping technology, relation language materials are generated in an iteration mode, and finally a relation corpus is formed; a relation classification model training stage, wherein term vectors and position embedding are combined to build a sentence vector matrix representation to serve as input of a network, the convolutional neural network is built, the network is trained by means of a back propagation algorithm, and a relation classification model is obtained; an enterprise entity relation extraction stage in a web page, wherein the web page is preprocessed by combining web page text extraction with a named entity identification technology, and then enterprise entity relation extraction is conducted on the preprocessed web page. By means of the method, not only the defects of an artificial feature method can be overcome, but also the enterprise entity relation can be extracted from the web page more accurately and efficiently.

A shared file computer system has multiple users, and two or more transactions established by the users may contend for the same resource. Such contention may result in a deadlock condition. To search for such deadlocks, the system maintains a lock wait matrix which lists which transactions are waiting on which other transactions. The matrix represents a hierarchical wait tree. A deadlock is present when the top node transaction in the tree also occurs lower in the tree. In such a case, either the top node transaction or any other transaction in the path or branch between the two occurrences of the top node transaction is cancelled to resolve the deadlock represented by this path. However if the intermediary node is cancelled to resolve this deadlock, a similar deadlock can occur in another path and must be resolved also. There can be thousands of paths in the tree to be searched for such a deadlock condition, and to optimize the deadlock search after the intermediary node is cancelled, transactions leading from this intermediary node are not examined. Nevertheless, all deadlock conditions are detected.

A total floor reaction force required correction amount by which an error between an observed value of a total floor reaction force acting on a mobile object 101 and a desired total floor reaction force approaches zero is converted to a spring displacement amount of a position / posture of a representative contact surface representative of ground surfaces of the mobile object 101. A correction amount of a displacement amount of each joint of the mobile object 101 is determined by multiplying the spring displacement amount by a pseudo inverse matrix of a Jacobian matrix representing a relation between a change amount of the position and posture of the representative contact surface per unit time and a change amount of a generalized variable vector per unit time. The displacement amount of each joint is controlled according to a corrected desired joint displacement amount obtained by correcting a desired joint displacement amount.

Computer aided visualization and diagnosis by volume analysis of optical coherence tomography (OCT) angiographic data. In one embodiment, such analysis comprises acquiring an OCT dataset using a processor in conjunction with an imaging system; evaluating the dataset, with the processor, for flow information using amplitude or phase information; generating a matrix of voxel values, with the processor, representing flow occurring in vessels in the volume of tissue; performing volume rendering of these values, the volume rendering comprising deriving three dimensional position and vector information of the vessels with the processor; displaying the volume rendering information on a computer monitor; and assessing the vascularity, vascular density, and vascular flow parameters as derived from the volume rendered images.

The invention discloses a car plate detection method based on a multiple feature and low rank matrix representation. The car plate detection method based on the multiple feature and low rank matrix representation includes: (1) inputting an image; (2) extracting color features and self-adaption LBP operator features; (3) obtaining a multiple feature and low rank matrix representation model; (4) dividing and solving the model so as to obtain a sub model; (5) outputting a fake car plate area and obtaining a final accurate car plate area; (6) correcting a car plate and outputting an image after being positioned. The car plate detection method based on the multiple feature and low rank matrix representation combines a plurality of features of the car plate, can effectively improve robustness and accuracy of car plate detection, reduces false detection rate, and can be used in an intelligent transportation system.

A system for adaptive partitioning of circuit components during simulating of a circuit having a hierarchical data structure includes a simulator module having one or more computer programs for 1) selecting a group of leaf circuits from the first branch and the second branch for simulation, where each leaf circuit is represented by a matrix comprising a set of equations, 2) determining a strength of coupling between two or more leaf circuits of the group in accordance with a set of predetermined electrical coupling criteria, 3) if two or more leaf circuits are deemed be strongly coupled, combining the corresponding matrix of each strongly coupled leaf circuit into a combined matrix, and 4) performing computation for the two or more strongly coupled leaf circuits in accordance with the combined matrix. The system adaptively adjusts the group circuit matrix for computing a group of circuits according to the strength of coupling between the circuits. Hence, it achieves higher simulation performance by reducing either the size of the solver matrix when the circuits are loosely connected to each other, or by reducing the number of computational repetitions due to the communication of changes of signal conditions between circuits by combining the individual circuit matrices when such circuits are closely coupled to each other.

Method and apparatus of a multiple-access communication system for simultaneous communication with a plurality of user entities are disclosed. Signature sequences are used to discriminate between signals to and / or from various user entities. In an example matrix representation, signature-sequences are describable as elements of a column of a signature-sequence matrix, said signature-sequence matrix being a low-density matrix and a matrix wherein non-zero elements of any single row of the matrix constitute distinct elements of a finite set of elements.

A system and method are disclosed for waveform based variational static timing analysis. A circuit is divided into its linear circuit parts and non-linear circuit parts and modeled together, by a combination of linear modeling techniques, into linear equations that may be represented by matrices. The linear equations in matrix form may be readily solved by a computer such that an input waveform to an input pin of the circuit can be sequentially “pushed” through the various interconnects and logic networks of the circuit to an output pin. Output voltage waveforms are obtained at each stage of the waveform pushing and may be used to perform static timing analysis.

The invention discloses a method and a device for the on-line selection of seats at a cinema. The method comprises the steps of receiving an on-line seat selection instruction for the selection of seats in a designated movie hall at a designated cinema; acquiring a seat map of the designated movie hall according to the on-line seat selection instruction, wherein the seat map is represented in the form of a matrix; acquiring the first historical seat selection data of the designated cinema; determining a recommended first seat according to the first historical seat selection data; returning the first seat to a client and indicating the client to differentially display the first seat in the seat map. According to the technical scheme of the invention, the device is capable of selecting appropriate seats for users, so that users no longer need to manually select seats. The device simplifies the on-line seat selection operation and facilitates the usage process of users. Especially for users using small screens and users that are not familiar with movie halls, the hesitate mentality of users during the seat selection process is avoided. Meanwhile, the seat locking time is saved.

A new data structure and algorithms which offer at least equal performance in common sparse matrix tasks, and improved performance in many. This is applied to a word-document index to produce fast build and query times for document retrieval.

A method for rapid OCT image acquisition includes acquiring by OCT a plurality of compressive measurements (y) representing a set of under-sampled OCT data in a Dirac domain below a Nyquist rate by sampling an object of interest at randomly spaced vertical and horizontal lines in a Cartesian geometry using a raster scan, and recovering a 3D volumetric OCT image (f) from the compressive measurements (y) using compressive sampling. The method may also include recovering the 3D volumetric OCT image (f) from the compressive measurements (y) based at least in part on a sparsifying matrix (S) capable of transforming the 3D volumetric OCT image (f) into a sparse representation, such as a matrix representation of the 3D volumetric OCT image (f) in a shift-invariant wavelet transform domain. The method may also be applied to radial OCT scan patterns.

A method of two-way authentication between a user and a known host using a non-disclosing password entry system generates a matrix of characters having a random characteristic with random characteristics being selected from a set of custom symbols, pictures or patterns (rather than alpha-numeric characters) that only the user recognizes. When the user sets up an account with the known host, a subset of these characteristics is predetermined for use specifically by the user. One or more of these may additionally be used in the user's PIN or password for easy memorization, allowing the user to first authenticate the log-in screen before the user enters the PIN for user authentication to the known host. Alternatively, randomized alpha-numeric characters may be used, but with a predefined grouping or subset of the characters in a predefined position on the initial character matrix presentation. If the user doesn't see the predefined special characters or figures in the character matrix, or the particular alpha-numeric subset in the character matrix, then the log-in screen is recognized as a fake.

Spatial and temporal characteristics of received radio signals in a multiple user radio system can be estimated based on signals received from the users. In one embodiment, the invention includes measuring radio frequency signals received at different elements of an antenna system over time, the received signals corresponding at least in part to transmissions received from a system user, accumulating the signal measurements into a first matrix, and generating a second matrix representing the transmissions of the system user. The invention further includes cross-correlating the first and second matrices to form a first cross-correlation matrix, cross-correlating the first matrix with itself to form a second cross-correlation matrix, and multiplying the second cross-correlation matrix with a product of the first cross-correlation matrix to form a fifth matrix in which the elements of the fifth matrix characterize the radio channel traversed by transmissions of the system user.

Spatial and temporal characteristics of received radio signals in a multiple user radio system can be estimated based on signals received from the users. In one embodiment, the invention includes measuring radio frequency signals received at different elements of an antenna system over time, the received signals corresponding at least in part to transmissions received from a system user, accumulating the signal measurements into a first matrix, generating a second matrix representing the transmissions of the system user, cross-correlating the first and second matrices to form a third matrix, and multiplying the third matrix with a product of the second matrix to form a fourth matrix in which the elements of the fourth matrix characterize the radio channel traversed by transmissions of the system user.

The invention relates to a traffic flow prediction method of a deep network based on fusion of spatiotemporal features. The method comprises the following steps: step A, acquiring historical traffic flow data containing spatiotemporal information from a traffic platform; step B, preprocessing the historical traffic flow data containing the spatiotemporal information to obtain space-time matrix representation of the historical traffic flow data; step C, training a deep learning network TSNN by taking the space-time matrix representation as the input of the deep learning network; and step D, inputting a traffic flow data sequence to be predicted into the trained deep learning network to obtain a prediction result. According to the method, the problems of incomplete feature extraction and incomplete feature fusion in traffic flow prediction are solved, and the accuracy and the precision of traffic flow prediction are improved.