71 results about "Hierarchical decomposition" patented technology

A method and apparatus for clustering data, particularly regarding an image, that constructs a graph in which each node of the graph represents a pixel of the image, and every two nodes represent neighboring pixels associated by a coupling factor. Block pixels are selected with unselected neighboring pixels coupled with a selected block to form aggregates. The graph is coarsened recursively by performing iterated weighted aggregation to form larger blocks (aggregates) and obtain hierarchical decomposition of the image while forming a pyramid structure over the image. Saliency of segments is detected in the pyramid, and by computing recursively, a degree of attachment of every pixel to each of the blocks in the pyramid. The pyramid is scanned from coarse to fine starting at the level a segment is detected, to lower levels and rebuilding the pyramid before continuing to the next higher level. Relaxation sweeps sharpen the boundaries of a segment.

The present invention is a method and apparatus for automatically generating an optimal configuration of a product having a complex design, using logic implemented on digital computer processing system. A general configuration for the product will be broken down into a hierarchy of subdesigns by a designer of an artifact type, such as a bus. A particular instance of the product type must satisfy a set of external parametric constraints specified by a user. These constraints may take the form of a range of values for some performance characteristic of the product. Constraints may also be imposed for other reasons, such as governmental regulations or business requirements. The hierarchical decomposition facilitates the solution of highly complex problems. Criteria for a best solution may be specified for a given subdesign, for a collection of subdesigns, or globally. The tentative selection of a particular subdesign may impose internally generated parametric constraints upon a subsequent subdesign solution. If no acceptable solution is found for a given subdesign, the candidate overall configuration is rolled back to the most complete viable partial collection of subdesigns. The method transforms the constraints into a concrete design. The method will ordinarily be executed on a digital computer, and the instructions to program the computer will be stored on tangible digital media.

The invention relates to a safety and stability control system modeling method in power system dynamic simulation. The method comprises the following steps of: performing hierarchical modeling on the safety and stability control system in a hierarchical decomposition mode; simulating an operation mode, fault information and a control strategy by adopting a ladder logic diagram to obtain a main station model and a sub-station model; and connecting a safety and stability control system model and a primary system model to realize the close-loop dynamic simulation of a large-scale power system which contains the safety and stability control system model. By the method, various different action behaviors and control characteristics of the safety and stability control system in the large-scale power system can be simulated accurately; the defects that conventional simulation software cannot reflect the action behaviors of the safety and stability control system are overcome; postmortem analysis of complex and severe accidents can be performed; and the requirements of the large-scale power system dynamic simulation on the simulation of the safety and stability control system can be met.

A tree-structured index to multidimensional data is created using occurring patterns and clusters within the data which permit efficient search and retrieval strategies in a database of DNA profiles. A search engine utilizes hierarchical decomposition of the database by identifying clusters of similar DNA profiles and maps to parallel computer architecture, allowing scale up past previously feasible limits. Key benefits of the new method are logarithmic scale up and parallelization. These benefits are achieved by identification and utilization of occurring patterns and clusters within stored data. The patterns and clusters enable the stored data to be partitioned into subsets of roughly equal size. The method can be applied recursively, resulting in a database tree that is balanced, meaning that all paths or branches through the tree have roughly the same length. The method achieves high performance by exploiting the natural structure of the data in a manner that maintains balanced trees. Implementation of the method maps to parallel computer architectures, allowing scale up to very large databases.

A signal processing method for performing hierarchical empirical mode decomposition (H-EMD) and an apparatus therefor are provided. In an embodiment, when empirical mode decomposition is performed on an input signal, an artificial assisting signal and the input signal are combined to assist the search for extrema and frequency reduction is performed in each iteration to eliminate the artificial assisting signal and make mode decomposition convergent so as to avoid mode mixing. In addition, in an embodiment, a hierarchical decomposition method is provided to decompose the input signal into a fewer number of fundamental modes. For needs in application, one of the fundamental modes can be further decomposed to produce a number of supplementary modes. In an embodiment, the H-EMD with appropriate frequency reduction can result in modes substantially independent of the form or the way of envelopes and can be applied to decompose multi-dimensional signals.

An image interpolator for high-definition images is presented that utilizes a two or more level hierarchical decomposition for complexity reduction. At the lowest level, after a block-based motion estimator, a pixel-based motion vector selector and a motion vector-based halo reducer, an occlusion-based adaptive motion field interpolator provides preliminary motion vectors for higher resolution level. At the high resolution level, a new motion estimation refining based on the double consideration of eventual occlusion region and motion vector precision refines the preliminary motion vectors, and again a motion vector-based halo reduction re-corrects the refined motion vectors. A new motion compensated image interpolator with controllable features for multiple interpolated images is presented. A versatile post-processing configuration for remaining artifacts may also be included.

The invention discloses a space trajectory retrieval method of body movement data, comprising the following steps: definition of a body movement structural mode: performing movement correlation hierarchical decomposition to body joint mode to form the body movement structural mode comprising five substructures; generation of offline movement feature tree: to each movement in a body movement base,calculating the space trajectory curve of the joint relative to affiliated substructure root node joint and extracting the feature of the space trajectory curve as the movement feature of the joint; generating a movement feature tree comprising five movement feature sub-trees according to hierarchical definitions of the five substructures in the body action structural mode; retrieval of online movement data: submitting retrieval example and assigning retrieval joint by the user; obtaining a retrieval tree according to the assigned retrieval joint and the body movement structural mode; performing feature similarity calculations by the retrieval tree and the movement feature tree according to the hierarchical relationship from top to bottom in order, feeding back result movements after sorting the final similarities from high to low.

Provided is a mode-dependent micro grid multi-mode operation switching control method. The output characteristics of various distributed power supplies and the characteristics of signal transmission between the distributed power supplies are analyzed. A micro grid system is decomposed into a dynamic model of subsystems in a hierarchical decomposition approach according to the multi-path power flow of energy inside the micro grid. A networked control model of an inverting micro source and comprehensive state space static and dynamic models of the micro grid under various operation states are built based on a small signal stability method. The stability of the micro grid system in all states and at the moment of state switching and the switching control law for ensuring smooth transition between the states are studied based on a switching system theory. The stability of the micro grid system is well guaranteed at the moment of operation mode switching. The stability control problem and the energy balance and power balance tracking control problem in multi-mode operation of the micro grid are solved. Stable operation of the micro grid and seamless switching between modes are realized to the maximum.

Embodiments of the invention comprise a new device and method to realize an improved video frame memory reduction for a video decoder. In one embodiment, this improvement is achieved by a removal of the rate controller and the utilization of both a block compression technique and a fixed storage allocation technique, in order to lower the overall system cost, and to lower the frame memory requirements. In a preferred embodiment, this improvement is achieved by performing a hierarchical transform, for example, a Haar transform, that operates on the previously decoded frames. Then, the coefficients obtained from this transformation are quantized and then run-length coded, utilizing variable-length codes. The hierarchical transform preferably operates on an NxN block size with L levels of hierarchical decomposition, where N and L can be selected in advance. For example, in one preferred embodiment, N may equal 8, and L may equal 3. The compression system then fits the NxN block into an allocated storage of (NxN) / cf bytes, where cf designates the compression factor. For example, a nominal value of cf that equals 2, 3, or 4 may be utilized. The quantization process comprises a simple scaling of the coefficients. However, the DC coefficient is not scaled. The variable-length encoder comprises a run-length encoder that fits as many coefficients as is possible into the available space of the (NxN) / cf bytes.

The invention discloses a colorful printing mechanism of a 3D printer. The colorful printing mechanism of the 3D printer is composed of printing cooling fans, a fan support, a temperature control fusion extrusion component, a fixing plate, feeding pipes, four-color feeding motor devices, a colorful material extruding nozzle cooling fan and a colorful material extruding nozzle. The 3D printing mechanism further comprises a microprocessor control device, an X-axis moving device, a Y-axis moving device, an X-axis hierarchical moving module, a Y-axis hierarchical moving module, a 3D color module, a hierarchical decomposition module, a four-color segmentation value calculating module and a four-color feeding speed value generating module. By the adoption of the colorful printing mechanism, four standard colors are adopted, the four-color feeding motor devices control the feeding ratio of the four standard colors so as to achieve the purpose of making different colors through mixing, a user just needs to buy the consumables of cyan, magenta, yellow and black, and colorful objects in any shape can be directly generated from computer graphic data. Therefore, the production period of products is greatly shortened, the production efficiency is improved, and the manufacturing cost is lowered.

The invention relates to a method for optimizing an automobile product development system based on hierarchical decomposition. The method comprises: determining an optimized product development target; transmitting the development target in the entire vehicle hierarchy to lower hierarchies, namely the system hierarchy and the component hierarchy; realizing synergistic optimized solving among various hierarchies; and realizing the flexibility of a development task and modularized execution of development subtasks by combing the method with a multidisciplinary design optimization method. The invention provides the method which has quick convergence speed and high solving quality and can simultaneously realize optimization of development performance indexes and a design proposal in the entire vehicle hierarchy, the system hierarchy and the component hierarchy.

A method determines temporal patterns in data sequences. A hierarchical tree of nodes is constructed. Each node in the tree is associated with a composite hidden Markov model, in which the composite hidden Markov model has one independent path for each child node of a parent node of the hierarchical tree. The composite hidden Markov models are trained using training data sequences. The composite hidden Markov models associated with the nodes of the hierarchical tree are decomposed into a single final composite Markov model. The single final composite hidden Markov model can then be employed for determining temporal patterns in unknown data sequences.

The invention discloses an evaluation method for an implementation effect of city scape planning. The evaluation method comprises the following steps of 1) adopting an analytic hierarchy method, confirming evaluation indicators of top end scape planning for hierarchical decomposition, and constructing a multi-hierarchy evaluation system of scape planning, 2) comprehensively adopting a qualitative evaluation method, a quantitative evaluation method, a probability evaluation method and an ambiguous comprehensive evaluation method, and establishing an evaluating mathematical model according to hierarchies of the evaluation system of the landscape planning, 3) collecting evaluation data, inputting the evaluating mathematical model, and calculating a comprehensive evaluation value hierarchy by hierarchy, 4) establishing an evaluation analysis matrix, and analyzing and confirming evaluation grades, and (5) establishing an importance analysis matrix of the evaluation indicators, and analyzing and confirming importance values of the evaluation indicators in different hierarchies. The evaluation method for an implementation process and the implementation effect of the city scape planning combines an evaluation indicator system of the city scape planning with the evaluation matrices, so that the ambiguous implementation effect is quantified, and the evaluation method is applicable to evaluating the implementation effect of the city scape planning.

The invention discloses a method for quantifying surface appearance of a representation film based on multi-scale system theory. The method comprises the following steps: firstly using a scanning probe microscope to acquire a surface appearance image of the film through multistage conversion of scanning size, then judging whether the surface appearance of the film has multi-scale characteristics by using a multi-scale system analysis tool, and determining the characteristic size of a fluctuation structure on the surface of the film; then aiming at the surface appearance image with the multi-scale characteristics, using a two-dimensional wavelet packet analytic method to carry out multi-scale layer decomposition on the surface appearance image; comparing decomposed image group components with characteristic size values in size to determine the reconstruction delivery size corresponding to the decomposed image group components, and then selecting and reconstructing the image group components to obtain surface fluctuation structural image of the film with different characteristic sizes; and finally using a surface roughness method to evaluate a reconstructed image to acquire related quantitative information. The method is suitable for exquisite and perfect quantified representation for the surface appearance of a film material in micro / nano devices.

The invention discloses a distributed model predictive control method based on hierarchical decomposition, comprising the following steps: for a distributed system, an adjacency matrix of the distributed system is acquired according to a communication network structure, and each subsystem is divided into a plurality of connected sets by the adjacency matrix by using an adjacent matrix based path search method; a reachable matrix of the connected sets is constructed, the level of each connected set is determined based on a hierarchical decomposition method in an interpretative modeling method, all the connected sets of the same level are combined into one connected set, and therefore, a connected-set set of a serial structure is constructed; and at each sampling moment, the optimal control input sequence of each subsystem in each connected set is solved in sequence according to the series order, and predictive control is performed on the distributed system. By using the method, communication between connected sets not correlated directly is avoided, the communication burden is reduced greatly while the system stability is ensured, and the problem that the traditional collaborative distributed model predictive control method is of high necessary communication burden is solved.

The invention discloses a method for filtering online monitored noise of steel rail cracks and judging the cracks. The method comprises the following steps: collecting steel rail sound signals by virtue of an acoustic emission sensor, and transferring the steel rail sound signals back to a monitoring terminal by virtue of a collection controller, and processing; carrying out hierarchical decomposition on the steel rail sound signals by virtue of a db6 wavelet, so as to obtain a wavelet packet reconstruction signal component, time, frequency and a wavelet coefficient; calculating a frequency band energy ratio, so as to eliminate high-frequency interference in the wavelet packet reconstruction signal component caused by striking; reconstructing the wavelet coefficient of the rest wavelet packet reconstruction signal component, so as to obtain reconstruction signals; carrying out classical acoustic emission parameter analysis on the reconstruction signals, so as to obtain acoustic emission parameters; and analyzing correlation diagrams and experience diagrams, so as to judge whether the cracks are expanded. By virtue of the method, noise interference signals in acoustic emission signals can be effectively removed, crack signals can be separated from the interference signals, and the separation precision is high; and the method can be used for processing smooth signals and has a local analysis function to non-smooth signals.

The invention discloses a method and a device thereof for performing hierarchical empirical mode decomposition on a signal. In the embodiment of the invention, when the empirical mode decomposition is performed on the input signal, an artificial auxiliary signal is combined with the input signal in aid of searching an extreme value; and frequency reduction is performed in each iterative process so as to eliminate the artificial auxiliary signal and perform mode convergence at the same time to prevent mode mixing. Moreover, one embodiment of the invention provides a hierarchical decomposition method, which comprises the following steps: first, decomposing the signal into a relatively smaller number of intrinsic modes; and then expanding one intrinsic mode into expansion modes according to needs in application. In the embodiment, the mode decomposed by the hierarchical empirical mode decomposition under the proper frequency reduction is substantially irrelative to an envelope construction method; and the hierarchical empirical mode decomposition can be applied to the decomposition of a multi-dimensional signal.

The invention discloses a fully-automatic grouping method of WEB pages based on title separator, which comprises the following steps: separators in titles of WEB pages are fully used to automatically generate pages grouping trees, and pages are automatically classified into each group. The method provided by the invention can be applied into the field of network WEB analysis, and the core concept comprises: (1) visited-page information including titles is obtained by WEB behavior acquisition technology; (2) separators in title characters are used for segmenting titles into a plenty of key phrases; (3) a tree-like hierarchical structure corresponding to each key phrase is setup, and the routes by way of root nodes, intermediate nodes and leaf nodes of the tree-like hierarchical structure can be obtained through hierarchical decomposition; (4) if the key phrase of the title hierarchical decomposition can not be found in corresponding hierarchical nodes of the tree-like structure (only limited in the route), the key phrase is taken as attribute to establish a new node according to the rule of the tree-like structure so as to realize the establishment of the pages sorted tree-like structure.

The present invention is a method and apparatus for automatically generating an optimal configuration of a product having a complex design, using logic implemented on a digital computer processing system. A general configuration for the product will be broken down into a hierarchy of subdesigns by a designer of an artifact type, such as a bus. A particular instance of the product type must satisfy a set of external parametric constraints specified by a user. These constraints may take the form of a range of values for some performance characteristic of the product. Constraints may also be imposed for other reasons, such as governmental regulations or business requirements. The hierarchical decomposition facilitates the solution of highly complex problems. Criteria for a best solution may be specified for a given subdesign, for a collection of subdesigns, or globally. The tentative selection of a particular subdesign may impose internally generated parametric constraints upon a subsequent subdesign solution. If no acceptable solution is found for a given subdesign, the candidate overall configuration is rolled back to the most complete viable partial collection of subdesigns. The method transforms the constraints into a concrete design. The method will ordinarily be executed on a digital computer, and the instructions to program the computer will be stored on tangible digital media.

The invention relates to a method for testing the reliability of complex systems, comprising the following steps: establishing a system hierarchy of several levels by identifying components of the system; identifying and assigning failure modes to components; determining test procedures for failure modes; reliable information about the availability of technical systems with the least amount of tests can be obtained by the following steps: identifying a level for allocating availability targets to components establishing a hierarchical decomposition of the system; allocating availability targets to the components identified in the previous step by qualitative methods; calculating a projected availability of all components; identifying components for which availability targets are not met by said projected availability and selecting another test or increasing severity of tests already selected till availability targets are met; and calculating an overall availability of the system.

A method determines temporal patterns in data sequences. A hierarchical tree of nodes is constructed. Each node in the tree is associated with a composite hidden Markov model, in which the composite hidden Markov model has one independent path for each child node of a parent node of the hierarchical tree. The composite hidden Markov models are trained using training data sequences. The composite hidden Markov models associated with the nodes of the hierarchical tree are decomposed into a single final composite Markov model. The single final composite hidden Markov model can then be employed for determining temporal patterns in unknown data sequences.

The invention discloses a language description guided video timing sequence positioning method, which comprises the following steps of S1, extracting a multi-modal feature coding network of cross-modal features so as to learn cross-modal information of videos and natural languages, and obtaining multi-modal fusion representation of input languages and visual modals; s2, performing hierarchical decomposition on the cross-modal information by adopting a hierarchical tree structure strategy; step S3, employing progressive reinforcement learning mechanisms, correct credit allocation is provided through two task-oriented rewards; mutual promotion of different strategies in the tree structure is encouraged; according to the method, the decision process of human beings from coarse to fine is simulated through the progressive reinforcement learning framework based on the tree structure, complex action strategies can be effectively decomposed, the number of search steps is reduced while the search space is increased, and a more impressive result is obtained in a more reasonable mode.

An image interpolator for high-definition images is presented that utilizes a two or more level hierarchical decomposition for complexity reduction. At the lowest level, after a block-based motion estimator, a pixel-based motion vector selector and a motion vector-based halo reducer, an occlusion-based adaptive motion field interpolator provides preliminary motion vectors for higher resolution level. At the high resolution level, a new motion estimation refining based on the double consideration of eventual occlusion region and motion vector precision refines the preliminary motion vectors, and again a motion vector-based halo reduction re-corrects the refined motion vectors. A new motion compensated image interpolator with controllable features for multiple interpolated images is presented. A versatile post-processing configuration for remaining artifacts may also be included.

The invention discloses a neural network acceleration method based on multi-block FPGA cooperative processing, and relates to the field of neural network optimization. Establishment of neural networkacceleration board card, an SOC chip and an FPGA are arranged on the acceleration board card; the SOC chip comprises a ZYNQ chip; the ZYNQ chip is interconnected with each FPGA; A ZYNQ chip is based on the complexity of a network model of a neural network. delay requirements and throughput requirements; decomposing the parameters of the network model according to layers; and dividing the FPGA flowseries according to the hierarchy of parameter decomposition, issuing parameters to the FPGA of the corresponding flow series according to the hierarchy of parameter decomposition, and controlling the FPGA started by each flow series according to the neural network model until the FPGA of which the flow series is the last level completes data processing.

A tree-structured index to multidimensional data is created using naturally occurring patterns and clusters within the data which permit efficient search and retrieval strategies in a database of DNA profiles. A search engine utilizes hierarchical decomposition of the database by identifying clusters of similar DNA profiles and maps to parallel computer architecture, allowing scale up past to previously feasible limits. Key benefits of the new method are logarithmic scale up and parallelization. These benefits are achieved by identification and utilization of naturally occurring patterns and clusters within stored data. The patterns and clusters enable the stored data to be partitioned into subsets of roughly equal size. The method can be applied recursively, resulting in a database tree that is balanced, meaning that all paths or branches through the tree have roughly the same length. The method achieves high performance by exploiting the natural structure of the data in a manner that maintains balanced trees. Implementation of the method maps naturally to parallel computer architectures, allowing scale up to very large databases.

The invention relates to a permission configuration method and device, and a storage medium, and belongs to the technical field of computers. The method comprises the following steps that: receiving apermission application request sent from a user terminal, wherein the permission application request at least comprises user identification, operation permission and a target path; according to a stored permission configuration file, determining a target path permission list matched with the target path; when the target path permission list is not determined, carrying out hierarchical decomposition on the target path to obtain the parent directory path of the target path; according to the permission configuration file, determining the permission list matched with the parent directory path asthe target path permission list; and adding the user identification and the operation permission to the target path permission list. By use of the method, the problem of low permission configurationefficiency due to the artificial adding of the permission can be solved, and permission configuration efficiency can be improved.

The invention provides a complex document separating and organizing method and a complex document automatic-generating method. The complex document separating and organizing method comprises the following step: decomposing a complex document into a document outline, document data and document styles, wherein the document outline of the complex document is defined as macroscopic description of a document structure, is used for hierarchical decomposition, definition and management of the document according to document content, and is a uniform organization of the document data and a document style template, the document data of the complex document is used for organization and description of document metadata, and the document style of the complex document is used for organization and description of the document styles.

The invention discloses a visualized display method for energy efficiency index decomposition. The method is characterized by comprising the steps that 1, index system structure information and a memory data model are established, wherein the index system structure information comprises a hierarchical structure relation between indexes and properties of index nodes; 2, an index layout position is calculated according to the superior and subordinate hierarchical relation between the indexes, index content, a font size, the frame decoration width of the index notes and the width-height spacing of the index nodes; and 3, an index billboard is rendered according to the width and height of canvas, a font, a font size and index layout position information, and the index nodes and an interactive control are rendered to the canvas by the adoption of an HTML in combination with an SVG mode, wherein SVG components are used to display all static elements on the billboard, and HTML elements are used to realize dynamic interaction. The index hierarchical decomposition billboard provides a dynamic interaction method, and it is convenient for a user to browse index information layer by layer and level by level.

The invention provides a system acceleration verification test method based on a fuzzy analytic hierarchy process. The method comprises the following steps: adopting the fuzzy analytic hierarchy process to decompose a to-be-verified system task reliability total index R from top to bottom according to four levels of function, structure, mechanism and stress, and determining the relationship between the total index R and each stress; distributing the weight of the total index R according to the stress type; obtaining the reliability of the total index R distributed according to the stress type;and determining an acceleration test profile, counting a test scheme and a total acceleration factor, and carrying out environment tests of different stress types on the system. The system acceleration verification test method based on the fuzzy analytic hierarchy process has the advantages that a mathematical statistics and fault physics method is adopted, the problem of complex system reliability index verification is solved through an environment test easy to carry out, the test time is short, the cost is low, the sample size is small, a test design method is provided, and the test efficiency is improved. Therefore, a test scheme can be quickly determined, and system reliability indexes are verified.

The invention provides a low-pressure-ratio axial compressor model level performance calculation method. The method is carried out according to the following steps of 1, collecting parameters: collecting environment parameters, flow parameters, inlet parameters, outlet parameters and early-stage preparation parameters of an axial compressor model level; 2, calculating overall performance parameters of the three-stage axial compressor model level: an overall energy head and overall variable efficiency; 3, obtaining a proportion coefficient of each stage of energy head and a proportion coefficient of variable efficiency; and 4, performing hierarchical calculation on the level performance parameters of each stage: according to the proportion coefficient of each stage of variable efficiency and the proportion coefficient of each stage of energy head, performing hierarchical decomposition on the overall performance parameters of the three-stage axial compressor model level obtained in the step 2. Through a multi-stage overall test, the decomposition calculation of each stage and the numerical simulation are combined to obtain relatively accurate axial compressor model level single-stageperformance parameters capable of guiding industrial application.