582 results about "Quantification methods" patented technology

Methods and systems utilizing the data provided by a non-contrast-enhanced CAC scan that is left unused by the “whole-heart” Agatston or volume scores. Agatston and volume scores summarize overall coronary calcium burden, but do not show the number of vessels involved, the geographic distribution of the lesions, the size and shape of the individual lesions and the distance of the lesions from the coronary ostium. The methods and systems described herein extract and use the enhanced information provided by 3-D CAC scan data and significantly increases its clinical predictive value by providing vessel and lesion specific CAC scores which are superior to the whole-heart Agatston and volume scores in predicting obstructive Coronary artery disease (CAD).

The present invention provides an analytical element which is free from evaporation of a sample during measurement and therefore capable of quantifying a substrate using a very small amount of sample with high accuracy and which is free from scattering of the sample during and after the measurement and therefore hygienically excellent; and a measuring device and a substrate quantification method using the same. The analytical element comprises a cavity for accommodating a sample, a working electrode and a counter electrode exposed to an inside of the cavity, a reagent layer which comprises at least an oxidoreductase and is formed inside or in the vicinity of the cavity, an opening communicating with the cavity and a member covering the opening.

The invention discloses a text feature quantification method based on comentropy, a text feature quantification device based on comentropy, a text classification method and a text classification device. The text feature quantification method comprises the following steps that: the weight of each feature word in a document is calculated according to the word frequency of feature words in a text document and the comentropy distributed on different text classes; meanwhile, the inter-class distribution entropy of the feature words is calculated in different modes according to the unbalance performance of the scale of each class of a text set; in addition, the inverse document frequency is introduced as required according to the distribution features of each feature word in the text set; local word frequency factors are properly reduced, so that the weight distribution of each feature word in the document is reasonable; and the feature differences of different classes of texts are sufficiently reflected by generated document feature vectors. The text feature quantification device and the text classification device disclosed by the invention have a plurality of options or parameters; and the optimum text classification effect can be achieved through regulation. The text feature quantification method has the advantages that the text classification accuracy is improved, and the performance on different text sets is stable.

The invention discloses a method and a system for wireless concentrated collection and management of energy measurement data. The system comprises wireless measurement instruments, collectors, a concentrator, wireless GPRS communication modules, and a system management center. The system is characterized in that: the system adopts a quantification method, a timing method, a multi-channel redundancy communication technology, and a high efficiency forward error correction channel coding technology, such that power consumption is reduced, and reliability and collection success rate are ensured; different wireless interfaces and different communication protocols are adopted in the communication between each subsystems of the wireless concentrated collection and management system, different wireless transmission frequencies are adopted in different wireless interfaces, and a transparent data transmission method is adopted, such that wireless data communication reliability, system compatibility, system flexibility and system applicability are ensured. With the system and the method provided by the present invention, data information which is real-time, real and reliable is provided for management departments, and real-time monitoring, statistics gathering, and analyzing of measurement instrument operation status are realized, such that managing and administrating capacities of management departments are improved.

The invention discloses an image stylizing method capable of being operated in real time. The method comprises the following steps: forming the encoding part of a generation network by using a fire module proposed in Squeeze-Net and forming a decoding part of the generation network by using a de-convolution layer; forming a discrimination network by using conv1_1 to conv5_1 convolution layers in the VGG19; performing semi-supervised learning by constructing a generation confrontational network; subjecting the de-convolution layer to flattened decomposition, dimensionality reduction, and elimination of redundancy by using a convolution kernel low-rank decomposition method on the premise that the picture synthesis capability of the generation network is not changed, and subjecting the whole generation network to further acceleration and compression by using a network pruning and parameter quantification method. The image stylizing method applies the Fire module and the Flat-Deconv module to the generation network for the first time, and integrates multi-scale information during encoding and decoding. Thus, the generation network is made lightweight when a certain synthesis capability of the generation network is guaranteed, so as to be more suitable for mobile terminals.

The invention provides a power grid running safety risk quantification method. The method comprises establishing a risk quantification model, wherein the power grid value risk is equal to consequence value*probability value; determining the category of influence factor items required to participate in risk quantification estimation to obtain the weighted value among all influence factor sub-items contained by each screened influence factor item; selecting the corresponding influence factor sub-items from each influence factor items participating in risk quantification estimation, multiplying the selected influence factor sub-items to obtain a probability value, multiplying the probability value by a consequence value to obtain the power grid risk value during the occurring of the dangerous factors of the power grid, and obtaining the power grid grade according to the power grid risk value. By the aid of the method, high risk quantification scientificity and accuracy is provided, and risks can be quantified reasonably to be used as the object measurement basis for guiding actual power generation.

The invention discloses a channel quantification method for a multi-user MIMO system based on limited feedback and a device thereof, which belongs to the technical field of wireless transmission. The method comprises the following steps: according to a channel state message, from a codebook choosing a codeword which has the smallest angle with the codebook and calculating an quantification error; choosing a codeword vector which has the smallest angle with the quantification error from the codebook; projecting a channel vector to the space formed by the two chosen codeword vectors, calculating and obtaining a compound proportion factor; quantifying the compound proportion factor by a pre-generated quantification step set. A receiver device comprises a channel state message obtaining unit, a channel vector qualification unit, a quantification message feedback unit. A sender device comprises a feedback message reconstruction unit and a pre-coding unit. The method and the device of the invention can reduce the quantification error and improve the capacity performance of the system and has the advantages of small codebook and low complexity with same rate and capacity.

The invention discloses a method for generating a remote sensing image quick view under Linux environment and belongs to the technical field of remote sensing image processing. The method comprises that (1) an image of a band sequence (BSQ) format is generated to an image of a tagged image file (TIF) format and (2) a quick view file of a joint photographic experts group (JPEG) format is generated, wherein the step (1) comprises of a, linear mapping, b, pretreatment of decimal part of image data, c, cutting of image data in accordance with a threshold, d, histogram equalization and e, generation of an image file of the TIF format. The method has the advantages of being simple and easy to operate. The generated quick view can describe data information of a current wave band of a hyperspectral image accurately, the method is used for processing 8-position unsigned char type data converted from 32-position single precision floating point type data, and the quantification method has low noise.

The invention provides a new method for double quantifying the protein group marked on the basis of stable oxygen isotope <18>O, which is characterized in that: based on the fact that the N end peptide section and the C end peptide section are generally labeled, and by means of combining the <18>O chemical labeling and enzyme digestion, the quantifying strategy for two <18>O markings of the same sample can be accomplished. The method relates to the reductive alkylation and enzyme digestion of protein, dual-functional reagent modifying and <18>O marking of peptides, multi-dimensional liquid phase chromatogram separation and multistage Tandem Mass Spectrometry quantification and identification. A first mass spectrum peak area is used to quantify; and the identification of the protein is conducted through a second stage mass spectrum and by searching the database. The invention also provides a reagent box for facilitating the implementation of the method.

The invention provides an intelligent analysis system and method for magnetic flux leakage detection data in a pipeline. The process comprises the following steps: adopting time domain-based sparse sampling and KNN-softmax data complete set construction module, obtaining a complete magnetic flux leakage data set; a pipeline connection assembly discovery method based on combination of selective search and a convolutional neural network is adopted in the discovery module to obtain the accurate position of the weld joint; adopting an abnormal candidate region searching and identifying method based on a Lagrangian multiplication frame and multi-source magnetic flux leakage data fusion in the discovery model to find out defective magnetic flux leakage signals; Adopting a defect quantification method based on a random forest in a quantification module to obtain a defect size; a pipeline solution improved based on the ASME B31G standard is adopted in the solution module, and an evaluation result is output. An analysis method is provided from the overall perspective, and pretreatment, connection assembly detection and abnormity detection, defect size inversion and final maintenance decision are achieved.

The invention relates to a bispecific antibody or antibody mixture, and more specifically relates to the bispecific antibody or the antibody mixture having a common light chain, and a preparation method of the bispecific antibody or the antibody mixture. The invention also relates to nucleic acid molecules for coding the bispecific antibody or the antibody mixture, a recombinant vector containing the nucleic acid molecules and recombinant cells, and a detection and quantification method for the bispecific antibody or the antibody mixture. The bispecific antibody or the antibody mixture is prepared by a common light chain technology, the preparation method is simple and controllable, mismatching of the light chains in the bispecific antibody can be avoided, the antibody mixture can be expressed in same host cell, difficulty for culturing the mixed cell groups is reduced, and the amplification production is easily achieved.

The invention discloses a method and a system for establishing a smart city 'multi-specification-in-one 'evaluation system. The establishment method comprises the following steps: constructing a framework of a smart city multi-standard-in-one effect evaluation index system, and forming a first-level index and a second-level index; establishing a three-level index of a smart city 'multi-standard-in-one' evaluation index system; quantifying the third-level index through an index quantification method; determining the weights of the first-level index, the second-level index and the third-level index; and calculating the score value of the smart city'multi-specification-in-one', and defining the evaluation level of the 'multi-specification-in-one' effect of the smart city. The implementation system comprises an evaluation index management module, an evaluation input module, an evaluation output module and a user management module. According to the method, the blank of the evaluation methodfor the implementation effect of multi-specification integration of the smart city is filled, the research category and the research method in the field are expanded, and the method has foresight, scientificity and operability.

A quantification method and an imaging method are disclosed, capable of quantifying the margin feature, the cysts feature, the calcifications feature, the echoic feature and the heterogenesis feature of a tumor, and capable of imaging the margin feature, the cysts feature, the calcifications feature and the heterogenesis feature of a tumor. The quantification method and the imaging method calculate the moving variance of the gray scale of each of the pixel points based on the gradient value of the gray scale of these pixel points. Then, depending on the purpose of the quantification method or the imaging method, the maximum value, the minimum value, the mean value, and the standard deviation of the moving variance of the gray scale of these pixel points are calculated, respectively. At final, with the definition of the threshold value and the imaging rule, the above features of the tumor are quantified or imaged.

The present invention is related to an identification and/or quantification method of a biological (micro)organism or part of it (possibly present in a biological sample) by a detection of its nucleotide sequence among at least 4 other homologous sequences and comprising the steps of: possibly extracting original nucleotide sequences (1) from the (micro)organism; amplifying or copying with a unique pair of primer(s), at least part of original nucleotide sequences (1) into target nucleotide sequences (2) to be detected; possibly labelling said target nucleotide sequences (2); putting into contact the labelled target nucleotide sequences (2) with single stranded capture nucleotide sequences (3) bound by a single predetermined link to an insoluble solid support (4), preferably a non porous solid support, discriminating the binding of a target nucleotide sequence (2) specific of an organism or part of it by detecting, quantifying and/or recording a signal resulting from a hybridization by complementary base pairing between the target nucleotide sequence (2) and its corresponding capture nucleotide sequence (3), wherein said capture nucleotide sequence (3) being bound to the insoluble solid support (4) at a specific location according to an array, said array having a density of at least 4 different bound single stranded capture nucleotide sequences/cm2 of solid support surface and wherein the binding between the target nucleotide sequence and its corresponding capture nucleotide sequence forms (will result in) said signal at the expected location, the detection of a single signal allowing a discrimination of the target nucleotide sequence specific of an organism or part of it from homologous nucleotide sequences.

The invention provides a reproducible, objective quantification technique that reliably segments white matter structures. The technique receives a seed voxel within the white matter structure from an individual, determines thresholds and selection criteria, creates a binary mask based on the at least one threshold and the at least one selection criteria and calculates the boundary of the white matter structure based on the binary mask. A magnification factor is applied to each component of the eigenvectors of voxels. Boundary voxels are determined wherein each of the boundary voxels has a magnitude above a predetermined value and is located next to a voxel having a magnitude below the predetermined value. A vector is drawn from the seed voxel to a boundary voxel and the boundary voxels are connected together, thereby forming the region of interest within the connected boundary voxels.

The invention discloses a deep convolutional neural network-based inter-layer non-uniform K-means clustering fixed-point quantification method. The method includes the following steps that: step 1, a part of images of a deep convolutional neural network which can be correctly identified are selected, and feature mappings (Feature Map) generated in an identification process are extracted; step 2, inter-layer non-regular quantification is performed on the feature mappings in the deep convolutional neural network, and the maximum number of quantification bits of each layer of the convolutional network is determined with the precision of the model maintained; step 3, for each convolutional layer in the model, a K-means clustering algorithm is used to determine fixed-point values satisfying feature mapping distribution, the range of the fixed-point values is made to be located in a range which can be expressed by the maximum number of quantification bits, and the fixed-point values are adopted to represent values in the feature mappings and are stored in the form of indexes; and step 4, a neural network model fine tuning method is adopted perform fine tuning on the model, so that error caused by the quantification can be eliminated. With the inter-layer non-uniform K-means clustering fixed-point quantification method adopted, the storage cost of the feature mappings of the deep convolutional neural network can be greatly reduced with the precision of the model maintained. The method is innovative.

The invention discloses a vector quantization method. The core idea of the method is that classifying the vectors to be quantized in accordance with the jump value between different dimension coefficients of the vectors to be quantized, and quantizing the vectors to be quantized and the residual deviation of the classified fixed track vectors. The invention also provides a corresponding vector quantizer. The classification mode adopted in the invention can perform preliminary determination of vector dimension coefficient variation trend, so that the difference codebook corresponding to each type of vector has small variation range, and the capacity of codebook used in the following quantization procedure is effectively reduced under the prerequisite of assuring reproduction quality, thereby reducing the required quantization bit number.

The invention discloses a fuzzy-neural-network-based tea leaf appearance quality quantification method, which comprises the following steps of: (a) selecting a batch of representative samples, and carrying out sensory evaluation on the tea leaf samples by tea leaf tasters with national certificates so as to obtain the appearance grading values of the tea leaf samples; (b) obtaining the visible images of the appearances of tea leaves by adopting a computer vision technology, respectively extracting shape features and color features after carrying out pretreatment on the visible images, and carrying out principal component analysis on all extracted feature variables to obtain a group of uncorrelated new variables; (c) establishing a fuzzy-neural-network-based tea leaf appearance quality quantitative evaluation model, taking the front p principal component factors extracted in the step (b) as an input layer of a network, and taking the sensory evaluation grading value of the appearances of the tea leaves as a desired output of a fuzzy neural network model, wherein the fuzzy-neural-network-based tea leaf appearance quality quantitative evaluation model comprises an input layer, a fuzzy layer, a fuzzy rule calculation layer and an output layer; and (d) calculating the graded value of the appearance quality of unknown tea leaf samples by using the established fuzzy neural network model for tea leaf appearance quality.

The invention provides a convolutional neural network quantification method, which comprises the steps of training a full-precision model of a convolutional neural network to be quantified, and calculating standard deviation of weight and response distribution of each layer of the full-precision model; estimating scale factors of parameters and features of the full-precision model according to thestandard deviation and hyper-parameters of the weight and response distribution of each layer of the full-precision model; for the to-be-optimized convolutional neural network, establishing a quantization module containing scaling factor-based forward calculation and backward gradient propagation functions to obtain a corresponding quantization network; carrying out fine tuning training on the quantization network, and determining an optimal scale factor; and retraining the quantization network generated by the optimal scaling factor to obtain a final quantization neural network model. The invention further provides a convolutional neural network quantization device, a computer and a storage medium. According to the invention, the problems of complex realization and high calculation complexity of the existing model quantification method are improved.