44 results about "Structural correlation" patented technology

Structural health monitoring device with improved reliability and performances that is applied in selected locations on a structure. The structural health monitoring device includes a data acquisition, process and storage media, a direct and independent wireless connection system to a standard and globally interconnected telecom network, is uninterruptedly powered by a power management system featuring at least two battery power sources. The structural health monitoring device further includes sensors specifically intended to remain permanently active and “asynchronously” trigger data acquisition sessions in occurrence of timely unpredictable phenomena having structural relevance and is moreover comprising data processing media for data compression and for automatic detection of possible structural anomalies using a self-training neural data processing algorithm.

The invention discloses a video image character recognition method based on submesh characteristic adaptive weighting. Aiming at the technical problems in the conventional video image character recognition, the invention provides an image character binaryzation method based on edge characteristics. After a binary image of a character is acquired, the image character is subjected to regional mesh division by an elastic mesh method; and before the elastic mesh characteristic of the image character is calculated, the relevance of stroke structures of adjacent meshes is extracted. By the method, the stability and the robustness of extracting the characteristic of the character are improved, so that the recognition rate of video image characters is increased.

The invention discloses a traffic data make-up method. The traffic data make-up method comprises the following steps that S1, random deficiency processing is carried out on a complete traffic data set according to a set traffic data deficiency rate, and a complete traffic data set subjected to deficiency processing is obtained; S2, normalization processing is carried out on the complete traffic data set subjected to deficiency processing and the complete traffic data set, and normalized traffic data are obtained; S3, a traffic data make-up model based on a noise reduction pile type automatic encoder deep-layer network structure is trained, and a make-up model is built; S4, input incomplete traffic data containing missing data are made up for by calling the make-up model, and traffic data values obtained after make up are obtained. According to the traffic data make-up method, the missing data and observed data are treated as a whole, the traffic data are made up for on the aspect of data recovery, the structural correlation among the traffic data is deeply excavated, the missing data are made up for at one time, the efficiency is high, and robustness is good.

A hyperspectral image inpainting method based on E-3DTV regularity is provided. The method comprises the steps of expanding the original three-dimensional hyperspectral data with noise into matrix along the spectral dimension, and initializing the matrix representation of the noise term and the hyperspectral data to be repaired, and other model variables and parameters under the ADMM framework; performing differential operation on the hyperspectral data to be repaired along horizontal, vertical and spectral dimensions to obtain three gradient maps in different directions, which are expanded into matrices along the spectral dimensions; decomposing the gradient graph matrices in three directions by low rank UV, and constraining the basis matrices of the gradient graph by sparsity to obtain E-3DTV regularity; adding the E-3DTV regularity to the data to be repaired, writing out the optimization model, using the ADMM framework to solve iteratively; obtaining the restoration image and noisewhen the iteration is stable. The invention performs denoising and compression reconstruction on the hyperspectral image data, and improves the enhancement of the traditional 3DTV so as to give consideration to the structural correlation and sparsity of the gradient image, thereby overcoming the defect that the traditional 3DTV can only depict the sparsity of the gradient image and ignores the correlation.

An efficient conversion method for a deep learning model provided by the embodiment of the invention is used to solve the technical problem that the development efficiency and operation efficiency of a deep learning model are low. The method includes the following steps: building a data standardization framework corresponding to an NPU (Neural-Network Processing Unit) model according to a general deep learning framework; using the data standardization framework to convert the parameters of a deep learning model into the standard parameters of the data standardization framework; and converting the standard parameters into the parameters of the NPU model. According to the invention, a unified data standardization framework is built for a specific processor according to the parameter structures of general deep learning frameworks. Standard data can be formed using the unified data structure of the data standardization framework according to the parameters of a deep learning model formed by a general deep learning framework. Thus, the process of data analysis by the processor depends much less on the structure of the deep learning model, and the development of the processing process of the processor and the development of the deep learning model can be separated. A corresponding efficient conversion device is also provided.

The invention relates to a related prediction model-based method for detecting structural deformation in a magnetic resonance image, which is technically characterized in that: a group of normally tested triangular cerebral cortex surfaces which are registered to a standard template is utilized to calculate the structural dependence between the vertex on the group of surfaces and other vertexes; the structural dependence and typical related prediction models are utilized to predict an expected position of the vertex existing in a cerebral atrophy area according to a vertex position of a normal area without structural deformation of the brain; the vertex position obtained by the prediction models is compared with the vertex position before prediction to quantize the deformation on the cerebral cortex surface due to the cerebral atrophy so as to quantize the degree of the structural deformation of the brain resulted from the cerebral atrophy. Compared with other methods, the method has the main advantage that: the method can detect the presence of the structural deformation of the brain and quantize the degree of the structural deformation under the condition that only a single time-point magnetic resonance structure image is available.

The invention discloses a no-reference blurred image quality evaluation method, and belongs to the image processing technical field. A characteristic that compared with a sharp image, edge features of a blurred image change greatly is utilized; all step edges of the image is sought; appropriate portion edges based on contrast ratio and pixel distance is selected; and the sharpness degree of the edges is calculated and taken as the evaluation basis for image blurriness. The method provided in the invention highlights the influence of human eye-sensitive edge information portion of the image on the image quality, so that images of the blur types, such as Gaussian blur, defocus blur, motion blur and the like can be well evaluated, structural dependence of subjective evaluation is high, and the characteristic of a human eye vision system is accorded with.

The invention discloses a multi-energy-spectrum CT projection domain base material decomposition method and device based on deep learning, and the method comprises the steps: employing multi-energy-spectrum CT to collect the multi-energy-spectrum multi-color projection of a known calibration model body in a network training process; using multicolor projection under an energy spectrum to directlyreconstruct a mold body CT image, segmenting the mold body CT image into a plurality of base material images, and respectively solving line integrals of the base material images along each ray; designing a deep neural network for multi-color projection decomposition, taking multi-energy-spectrum multi-color projection of the calibration model body as network input, and taking line integral of a base material image of the calibration model body as an output label to complete training; in the network application process, inputting multi-energy-spectrum multi-color projection of a measured objectinto a neural network, and decomposing line integration of a multi-base material image; and reconstructing a multi-base material density image of the measured object through the line integrals. The method is good in anti-noise performance, has no strict requirement on the correlation between the calibration mold body and the morphological structure of the measured object, and does not need to measure energy spectrum information in advance.

The invention discloses a hyperspectral image feature extraction method based on 3-D wavelet transform and a sparse tensor. The hyperspectral image feature extraction method includes step (1), balancing influence of data on feature extraction distinguishing according to a data normalization method; step (2), extracting spectral domain and spatial domain features from the normalized data by 3-D discrete wavelet transform; step (3), keeping good structural dependence between the features by expressing wavelet transform features as a second-order feature tensor form; step (4), achieving feature sparsification according to a sparse tensor distinguishing method; step (5), re-expressing the features subjected to sparsification as a vector form. By the hyperspectral image feature extraction method, classification accuracy of a whole classification system can be improved effectively.

The invention discloses a fabric thickness directional permeability testing device. The fabric thickness directional permeability testing device comprises an upper die, a lower die and a die cavity thickness support. The invention also discloses a testing method of a fabric thickness directional permeability. The method comprises the following steps of: a preparation step before injection; an injection step; and a step of measuring test data, and calculating the permeability. According to the invention, the influence of multilevel structure, multiple type unit cells and arrangement and combination manners of a prefabricated part on the spatial distribution of the prefabricated part permeability is analyzed, and a mathematical model of the correlation between the permeability and the distribution thereof and structure can be constructed, the forming mechanism and rule of the soakage defect of the composite material are revealed, and a criteria and a control method for the formation of the soakage defect are developed, so that the application and popularization of the composite material in the fields such as spaceflight and aerospace, automobile, building and wind power generation in China are promoted.

The invention discloses a device for testing intrinsic permeability of a radial surface of fiber fabric. The device comprises an upper die, a lower die, a die cavity thickness bracket and a piezoelectric sensor. The invention also discloses a method for testing the intrinsic permeability of the radial surface of the fiber fabric. The method comprises the following steps: preparation before injection; injection; measurement of test data; and calculation of the permeability. According to the device and the method, the influences of a multi-layer structure of a prefabricated member, multi-type cells and the permutation-combination mode of the multi-type cells on the distribution in space of the permeability of the prefabricated member are analyzed as key points, a mathematical model with the permeability and the distribution of the permeability being related with the structure is established, the formation mechanism and rules of a permeability defect of a composite material is disclosed, a formation criterion and a control method of the permeability defect are developed and therefore the application and the popularization of the composite material in the fields of aerospace, automobiles, buildings, wind electricity and the like are promoted.

A channel decoding method is provided. The method includes storing, in a memory, a set of first log likelihood ratio (LLR) values corresponding to bits of a codeword generated by modulation of a channel-encoded signal; changing, into a preset value, at least one LLR value corresponding to previously defined bits of the codeword from among the set of the first LLR values, to generate a set of second LLR values; and performing forward error correction (FEC) based on the set of the second LLR values and an FEC code, to estimate the bits of the codeword, in which the FEC code comprises a constraint code for constricting a previously defined structural correlation between the bits of the codeword.

Systems and methods for generating a magnetic resonance (MR) image of a tissue are provided. A method includes acquiring MR raw data. The MR raw data corresponds to MR signals obtained at undersampled q-space locations for a plurality of q-space locations that is less than an entirety of the q-space locations and the MR signals at the q-space locations represent the three dimensional displacement distribution of the spins in the imaging voxel. The method also includes performing a joint image reconstruction technique on the MR raw data to exploit structural correlations in the MR signals to obtain a series of accelerated MR images and performing, for each image pixel in each accelerated MR image of the series of accelerated MR images, a compressed sensing reconstruction technique to exploit q-space signal sparsity to identify a plurality of diffusion maps.

The invention relates to an encoding method of a light image with a color structure, belonging to the technical field of computer image processing. The encoding method of a light image with a color structure is designed by adopting the relevance of a topological structure and an internal structure, and the relevance among different atom areas and in the atom areas of a light encoding image obtained by shooting with the color structure is stronger, which is beneficial to the correct decoding of the light encoding image with the color structure. The light encoding image with the color structure can be projected on any object and has no strict limit to a curved surface for receiving the light encoding image with the color structure. The characteristic information of the whole space can be obtained by using one light encoding image with the color structure, thus the invention has strong real time and high identification precision rate.

The invention provides a multi-mode pattern matching method based on structure correlation against the multi-mode pattern matching problem. The redundant calculation problem can be solved. On the premise of adding a small amount of extra space based on the structure correlation of mode patterns, the overall performance of the pattern mode matching is improved. According to the method, a structureinheritance relationship between mode patterns is established mainly through structure correlation between the mode patterns, the originally mutually independent mode patterns are expressed as an integral data structure, and the data structure retains non-repeating parts which all the mode patterns need to be matched with. The mode pattern matching sequence is planned again, so that when the matching is performed in a data map, multi-time matching with the structurally repeated parts in the mode patterns can be avoided, so that the pattern mode matching time performance is greatly improved, and the goal of speeding up the calculation is achieved.

A system, method and apparatus for speckle noise removal based upon structural correlation in an OCT imaging system. In accordance with the present invention, several two- or three-dimensional OCT image scans can be acquired for processing by an adaptive algorithm. Specifically, at each image point an image intensity can be computed that quantifies a measure of dispersion of the image values in a particular direction with respect to their mean. Subsequently, a direction θ0(x, y) can be selected which minimizes the energy function at the given pixel (x, y). Finally, a value proportional to a local average of the input image around the point (x, y) can be chosen for the output image. In this way speckle noise can be minimized if not removed while, at the same time, maintaining the image substantially free of obvious artifacts.

The invention discloses an adaptive screening method for similar images of a capsule endoscope. The method comprises: all original images are divided into a plurality of sub image sequences according to a time sequence; and sampling is carried out on the sub image sequences, a similarity degree of has two adjacent images in a sampling sequence is calculated, similar images are screened out, and then the rest of image as a screened image is outputted. Compared with the prior art, the provided method has the following beneficial effects: temporal correlation and structural correlation of similar images of the capsule endoscope are considered fully and a structural similarity degree is used as a measure for an image similarity degree of the capsule endoscope; and thus compared with other methods, the provided method has high accuracy and the magnitude order of the image of the capsule endoscope is reduced, so that a doctor can carry out diagnosis and analysis conveniently.

The present invention relates to the field of video image processing technology, in particular to a new method based on the related real-time video target tracking. In the method, a target is tracked by the mode of regional image search, and an image quality parameter is used as the characteristic parameter of target tracking. The image quality parameter characterizes the similarity between images, so that the value of the image quality parameter can represent the success degree of the target tracking. The image quality parameter is composed of three parts, that is, a correlation coefficient, contrast distortion and brightness distortion. The correlation coefficient characterizes the structural correlation between two images; the contrast distortion characterizes the similarity degree of the two images in terms of contrast; the brightness distortion characterizes the similarity degree of the two images in terms of brightness. Consequently, the image quality parameter comprehensively considers the integral information of images and therefore is particularly used as the characteristic parameter of target tracking. Meanwhile, the present invention puts forward the corresponding update policy of threshold templates and increases the robustness of algorithms.

The invention discloses a damping device of a packaging box, which comprises a connecting piece and at least one damping piece. The connecting piece is provided with a first side wall and a second side wall which are opposite, and a first fixing part is arranged on the connecting piece. The damping piece is arranged between the first side wall and the second side wall, a first fixing groove is arranged on the damping piece, the first fixing groove corresponds to the first fixing part so as to embed the fist fixing part and fix the damping piece on the connecting piece. The damping device of the packaging box is used for enhancing the structural correlation between a paper box and a damping material, improving the utilization rate of the damping space, and further reducing the probability of article damage when the paper box is compacted, the consumption of paper box materials and operations steps of assembling the paper box.

The invention discloses a parallel acceleration method for multi-mode graph matching. The method comprises the following steps: 1) generating a multi-mode graph index of a mode graph set of a target domain; 2) adopting a layer-by-layer grouping strategy for the multi-mode graph index, namely evaluating the mode graph appearing in each layer in the multi-mode graph index to obtain the matching costof mode graphs in each layer, and then grouping the mode graphs of the layer according to the matching cost; and 3) respectively allocating threads to different groups and simultaneously carrying outmatching calculation. According to the method, the structural correlation among the pattern graphs is mined by adopting a Pattern Tree index construction algorithm, and a parallel matching strategy is designed for the pattern graph with relatively weak structural correlation, so that the matching performance is further improved.

The invention discloses a multi-scale structure relevance based pedestrian target identification method. First, saliency bottom layer characteristics of a target are extracted in multiple scales according to different validness of the target bottom layer visual characteristics of different scales. Seconds, according to the geometric structure consistence of the same kind of targets in different scales, a local structure mode of different scales is constructed by using a local restriction uniform encoding method on the target characteristic vectors in different visual characteristic channels. Finally, according to the characteristic dimension difference of the target in different scales, the local structure characteristics of the target in different scales are converted to a characteristicsub space having identical geometric structural features so as to improve the pedestrian target identification performance. The multi-scale structure relevance based pedestrian target identification method provided by the invention is prior to related alike methods internationally and has especially distinctive performance in classification identification of pedestrian targets having comparativelylarge resolution ratio difference in a monitoring video.

The invention relates to the technical field of aero-engine repair, in particular to a multi-life-piece replacement policy search algorithm which considers structural correlation. Based on comprehensive consideration of economic correlation and the structural correlation among aero-engine life pieces, and with making the replacement total cost of the life pieces the lowest in a whole life cycle being an optimizing target, a multi-life-piece opportunity replacement problem optimization model is established; aiming at characteristics of the optimization model, a four-model solution space reduction rule is provided, the search algorithm based on the reduction rule is provided based on the provided rule, and the algorithm can obtain an optimal solution of the model. Finally, through the adoption of a numerical number experiment and an application case, the provided algorithm is evaluated and verified. The result shows that the provided algorithm can achieve precise solution of a small-scale multi-life-piece opportunity replacement problem.

The invention belongs to the technical field of image editing, in particular to a method for superdividing stereoscopic images. The method of the invention comprises the following steps of: the position information of the left and right pictures is estimated by adopting an optical flow estimation network; high-resolution image is reconstructed by image super-division network; when reconstructing the left image in a stereoscopic image, not only the local redundancy and structural correlation of the pixels in the image are utilized, combined with the characteristics of correlation and complementarity between the left and right images of the stereoscopic image, more high-frequency information lost in the low-resolution left image is obtained to recover the content of the original image, and the effect of enhancing the rich detail texture of the image resolution is achieved. Experimental results show that this method can effectively improve the resolution of stereoscopic images, and make stereoscopic images have clearer visual quality, richer content and higher research and application value.

The invention discloses a multi-mode graph index construction method and system for weak structure correlation. The method comprises the following steps: 1) reading mode diagrams in a mode diagram setof a target field and generating a mode diagram mark ID for each mode diagram; 2) constructing a pattern graph isomorphism tree: judging every two pattern graphs, and if a sub-graph isomorphism relationship exists between the two pattern graphs, adding a pattern graph isomorphism tree pointing to the pattern graph with the larger scale from the pattern graph with the smaller scale to obtain the pattern graph isomorphism tree of the pattern graph set; 3) performing frequent subgraph mining on the pattern graph isomorphic tree, finding a public pattern graph and adding the public pattern graphinto the pattern graph isomorphic tree; 4) when one sub-pattern graph in the pattern graph isomorphism tree has a plurality of parent pattern graphs, reserving a unique parent pattern graph for the sub-pattern graph, and 5) calculating a minimum spanning tree of the pattern graph isomorphism tree, and carrying out depth-first traversal on the minimum spanning tree to obtain an optimal matching sequence of the pattern graph set.

The invention relates to an industrial energy calculation system. A process data acquisition module stores industrial process data acquired in a production process control system into a data memory; the data memory inputs the industrial process data into a data structural correlation database and a calculation module respectively; the data structural correlation database calculates corresponding intermediate data according to the industrial process data and inputs the intermediate data into the calculation module; the calculation module outputs corresponding enthalpy and entropy according to the intermediate data and industrial process data. The industrial energy calculation system outputs the intermediate calculation results by building the data structural correlation database and calculates in different energy calculation models, greatly simplifies the calculation process of energy data, is widely applicable to various different industrial enterprises, meets all kinds of requirements of enterprise energy management work and provides data basis for developing deep energy use diagnosis and analysis work for enterprises.

The invention discloses a traffic data make-up method. The traffic data make-up method comprises the following steps that S1, random deficiency processing is carried out on a complete traffic data set according to a set traffic data deficiency rate, and a complete traffic data set subjected to deficiency processing is obtained; S2, normalization processing is carried out on the complete traffic data set subjected to deficiency processing and the complete traffic data set, and normalized traffic data are obtained; S3, a traffic data make-up model based on a noise reduction pile type automatic encoder deep-layer network structure is trained, and a make-up model is built; S4, input incomplete traffic data containing missing data are made up for by calling the make-up model, and traffic data values obtained after make up are obtained. According to the traffic data make-up method, the missing data and observed data are treated as a whole, the traffic data are made up for on the aspect of data recovery, the structural correlation among the traffic data is deeply excavated, the missing data are made up for at one time, the efficiency is high, and robustness is good.

The invention relates to an encoding method of a light image with a color structure, belonging to the technical field of computer image processing. The encoding method of a light image with a color structure is designed by adopting the relevance of a topological structure and an internal structure, and the relevance among different atom areas and in the atom areas of a light encoding image obtained by shooting with the color structure is stronger, which is beneficial to the correct decoding of the light encoding image with the color structure. The light encoding image with the color structurecan be projected on any object and has no strict limit to a curved surface for receiving the light encoding image with the color structure. The characteristic information of the whole space can be obtained by using one light encoding image with the color structure, thus the invention has strong real time and high identification precision rate.

The invention discloses a fabric one-way surface permeability testing device. The fabric one-way surface permeability testing device comprises an upper die, a lower die, a die cavity thickness support and a piezoelectric transducer. The invention also discloses a testing method of a fabric one-way surface permeability. The method comprises the following steps of: carrying out a preparation step before injection; carrying out an injection step; and measuring test data, and calculating permeability. According to the invention, the testing method can selectively analyze the influence of a multilevel structure, multiple type unit cells and permutation and combination manners on prefabricated part permeability spatial distribution, and can establish a mathematical model of the permeability as well as relevance of distribution and structure; and composite material infiltration defect formation mechanism and rule are revealed, infiltration defect formation criterion and control method are developed, thus the composite material is promoted to be applied and popularized in the fields of aerospace, automobile, construction and wind power in China.