156 results about "Characteristic distribution" patented technology

The invention provides a voiceprint identification method based on a Gauss mixing model and a system thereof. The method comprises the following steps: voice signal acquisition; voice signal pretreatment; voice signal characteristic parameter extraction: employing a Mel Frequency Cepstrum Coefficient (MFCC), wherein an order number of the MFCC usually is 12-16; model training: employing an EM algorithm to train a Gauss mixing model (GMM) for a voice signal characteristic parameter of a speaker, wherein a k-means algorithm is selected as a parameter initialization method of the model; voiceprint identification: comparing a collected voice signal characteristic parameter to be identified with an established speaker voice model, carrying out determination according to a maximum posterior probability method, and if a corresponding speaker model enables a speaker voice characteristic vector X to be identified to has maximum posterior probability, identifying the speaker. According to the method, the Gauss mixing model based on probability statistics is employed, characteristic distribution of the speaker in characteristic space can be reflected well, a probability density function is common, a parameter in the model is easy to estimate and train, and the method has good identification performance and anti-noise capability.

A multispectral nevoscope that uses specific wavelengths in the visible and infrared spectrum of electromagnetic radiation to transilluminate a skin-lesion or a biological tissue or specimen for imaging and maps multispectral 2-dimensional images into 3-dimensional virtual space for providing 3-D distributions of pre-defined parameters representing the characteristic properties (such as melanin, hemoglobin and deoxyhemoglobin, etc.) of a skin lesion. Methods are disclosed for analyzing and using the characteristic distributions of specific parameters for detection and management of skin-cancers, or characterization of a biological tissue or specimen.

There is provided an illumination apparatus including an optical integrator including multiple optical systems for receiving light from a light source, and for producing multiple beams for illuminating a plane of mask or reticle on which a pattern is draw, a position of an incidence plane of the optical integrator and a position of the plane to be illuminated being arranged optically conjugate with each other, and a device for forming an optical characteristic distribution of light at the incidence plane, the light entering at least part of the multiple optical systems, whereby different areas on the plane of plane to be illuminated can be illuminated under different illumination conditions.

The invention provides a method for automatically detecting clouds interfere naval vessel target by an optical remote sensing image. The method can significantly improve the accuracy of detecting the naval vessel target shielded by clouds, and reduce omission factor and false alarm rate. The method is realized by the following technical proposal that: an adaptive threshold value of image region segmentation is evaluated according to image background and gamma characteristic distribution different of the naval vessel target; background region of ocean and clouds and a target region connected with the naval vessel and clouds thereof in the image are separated; a contour directional chain of a region of interest ROI can be obtained by a contour tracing method so as to strengthen pattern feature description of the target; the region contour directional chain is subjected to match searching by a template of the target contour directional chain so as to distinguish a naval vessel region and a cloud region; and multi-feature parameters of the naval vessel are extracted by a morphological and texture analysis method, and target determination is determined by an iterative decision-making method of multi-feature priority, thereby realizing naval vessel automatic detection of the remote sensing image under clouds interference.

A method for efficiently extracting a variation distribution of a characteristic for a semiconductor integrated circuit. The method extracts a characteristic distribution of a semiconductor integrated circuit by performing a mathematical analysis using a polynomial expression based on a variation distribution of a process sensitivity parameter.

The invention discloses a characteristic value Gaussian statistical property-based polarized synthetic aperture radar (SAR) image classification method, which mainly solves the problems that the prior art is insufficient on cognition of characteristic distribution properties and the category judgment limit needs man-made determination. The method comprises the following steps of: 1) performing characteristic value decomposition on all pixel points of polarized SAR images to be classified; 2) selecting different homogeneous regions as the most basic category representative regions, and extracting characteristic values for representing the homogeneous regions; 3) estimating Gaussian hybrid model parameters of the characteristic values lambda 1, lambda 2 and lambda 3 of the homogeneous regions by adopting an expectation-maximization (EM) algorithm respectively, and solving a probability density distribution function of the characteristic values; 4) solving a joint probability distribution function of three characteristic values of each homogeneous region; and 5) performing Bayesian classification on the pixel points in the homogeneous regions, and outputting the classification results. The method has the advantage of remarkable classification effect on the polarized SAR images, and can be used for target detection and target identification of the polarized SAR images.

The invention relates to a method and system for analyzing pipe rainwater and sewage mixed collection distribution. The method comprises the steps of S1, grouping pipes and installing a liquid level meter; S2, screening inspection well water level data; S3, prejudging whether pipe sections are subjected to mixed collection of rainwater and sewage or not; S4, judging whether the pipe sections are subjected to mixed collection or not, and if yes, judging the mixed collection position. According to the method, by arranging the liquid level meter in an inspection well and acquiring liquid level data for a long term, through the combination of hydraulic calculation and analysis and big data analysis, the status quo of whether the sewage pipes are pipes subjected to mixed collection of the rainwater or not and area characteristic distribution are obtained; moreover, through liquid level data, the data is processed, through hydraulic power, flow is calculated to judge whether the phenomenon of mixed collection of the rainwater and the sewage exists among pipe sections in a sewage pipe network or not, the distribution of the pipe sections subjected to mixed collection is obtained, the specific positions of the pipe sections subjected to mixed collection of the rainwater and the sewage are judged in a more economic and safer mode, and data support is provided for later pipe updating andimproving.

The invention provides a remote sensing type urban image extracting method. The method comprises the following steps: S1, performing characteristics extracting for altitude data of an ASTER VNIR satellite remote sensing image and derived products and a sample of a PALSAR HH / HV satellite remote sensing image; S2, extracting obvious urban and non-urban points based on the spectral characteristics of the urban and non-urban part; S3, performing confidence spreading by the LLGC using the obvious urban and non-urban points as the initial information based on the characteristics distribution feature of data to be classified, so as to obtain an urban confidence map; S4, obtaining the confidence of the whole remote sensing image, weighting and randomly sampling to obtain a training sample; S5, classifying the urban based on SVM, namely, classifying by the SVM method on the basis of the characteristic vectors extracted in the step S1 and the sample data extracted in step S4, and then obtaining a urban map subjected to binarization according to a classification label. With the adoption of the method, the problems of high cost and high time consumption and the like caused by manual sampling in the prior art can be solved.

The invention discloses a whole-network abnormal flow identification method based on flow characteristic distribution. The method comprises the following steps: firstly performing coarse-grained abnormality identification on network data flow in the whole network, extracting node pairs having abnormal flow from the whole network, determining a characteristic category of the abnormal flow, then performing fine-grained abnormality identification based on the extracted characteristic category and the abnormal nodes, determining an abnormal flow characteristic value, and finally collecting an abnormal flow set from the abnormal nodes based on the determined abnormal flow characteristic value. According to the method disclosed by the invention, the abnormal flow identification is performed on the network data flow in the whole network by adopting a hierarchical method, thereby not only improving the accuracy, but also bringing no huge measurement and calculation workload, the abnormal nodes in the whole network are accurately located, and the IP address, the port number and other flow characteristic values of the abnormal flow in the normal node pairs are obtained, and by adoption of the method disclosed by the invention, the abnormal flow can be classified so as to judge the reasons causing the network abnormality.

The invention relates to a monitoring method for real-time anti-cheating testing. The method comprises the steps of receiving a real-time log file, and obtaining time sequence data and real-time characteristic data after preprocessing; performing pre-estimation on the time sequence data, and detecting a time sequence abnormity; and performing distribution abnormity detection on the real-time characteristic data, and generating a monitoring strategy according to the time sequence abnormity detection result. According to the monitoring method, the monitoring system and the computing equipment, coordination of two detections of time sequence pre-estimation and characteristic distribution abnormity can be realized, thereby realizing two effects of real-time abnormity recalling and strategy monitoring early-warning.

A finite element simulation model for simulating real human brain electrical characteristic distribution is obtained through simulation of a human brain MRI or CT image. The model comprises a scalp layer, a skull layer, a cerebrospinal fluid layer, a pallium layer, an alba layer, a ventricle structure and an eyeball structure composed of a sclera structure and a vitreous body structure, wherein the ventricle structure is an independent structure, and therefore follow-up operation is simplified. The electrical conductivity and relative dielectric constant of the finite element simulation model are identical with those of the real human brain tissue and can be selected from an NIREMF public database according to actual requirements within a magnetic field frequency band of 10 GHz - 100 GHz, and the numerical values of the electrical conductivity and relative dielectric constant are assigned to the finite element simulation model by means of finite element simulation software. By the adoption of the finite element simulation model, real induced electromagnetic field distribution generated in a human brain under magnetic stimulation can be reflected, and the precision and reliability of simulation study are improved greatly.

A total random number sequence generator generates a total random number sequence of an entire circuit, as fabrication variation. A signal path random number sequence extracting section extracts, from the total random number sequence, a signal path random number sequence for a partial circuit obtained by dividing the entire circuit. A circuit simulating section executes Monte Carlo analysis using the signal path random number sequence for each partial circuit, thereby obtaining a desired circuit characteristic distribution. In this manner, correlation is maintained between divided circuit characteristic distributions and, in addition, the obtained circuit characteristic distribution is used for clock skew distribution calculation and others. Moreover, the circuit scale of a target of circuit simulation is reduced.

The invention discloses a crystalline lens biomechanics and optical property noninvasive in-vivo imaging system and a measuring method. The ultrasonic load system provides mechanical waves capable of enabling a crystalline lens to deform, a spectral domain OCT system achieves the structure and elastic imaging of the crystalline lens; on the basis of the yoke eliminating technology, OCT full-range imaging is achieved, full anterior segment two-dimensional or three-dimensional structure images from the anterior corneal surface to the retrolental surface, morphological and optical characteristic parameters of the crystalline lens are obtained through calculation, and a crystalline lens topographic map is constructed. Meanwhile, the phase sensitive OCT technology is utilized, strain distribution in the depth direction of the crystalline lens is obtained, and a two-dimensional or three-dimensional biomechanical characteristic distribution diagram of the crystalline lens is obtained through rebuilding. High-resolution imaging and precision measurement of biomechanics and optical properties of the crystalline lens provide the more comprehensive and complete imaging surveying basis for fundamental research, early diagnosis, operation plan formulating and postoperation result evaluation of cataract and other crystalline lens diseases, and the important practical significance and clinical significance are achieved.

The invention relates to a method for simulating an underwater polarization field of wave water surface transmitted light. The method comprises the following steps: firstly, building a polarization model based on Rayleigh scattering sky light, and calculating the polarization of incident sky light; according to different refraction coefficients corresponding to a parallel component and a vertical component of an incident light E vector, obtaining a Mueller matrix of a refraction process of a calm water surface; secondly, building a probability distribution model of a wave water surface element gradient, unifying the reference coordinate systems of the incident light and transmitted light, considering the shield effect among wind waves, and obtaining a Mueller matrix of a refraction process of a wave water surface; and finally, combining the Stokes parameter representation method to obtain a polarization field distribution mode of underwater transmitted light. The method can successfully simulate polarization characteristic distribution of underwater transmitted light of different sun positions and takes the influence of wind speed and wind direction on underwater light field polarization characteristic distribution into consideration. Meanwhile, the method can reveal polarization distribution rules of underwater transmitted light in a Snell window.

The invention relates to an electrical tomography regularization reconstruction method based on an adjacent point variation sum. The electrical tomography problem is regarded as a linear ill-posed problem, A is a sensitivity matrix, b is a relative boundary measured value vector, x is an imaging gray value vector corresponding to field domain substance electrical characteristic distribution and is called a solution vector, and the method is characterized in that the sum of absolute values of electrical characteristic quantity difference of all adjacent pixel points is used as a regularization function, and a Newton-Raphson iterative method is utilized to solve x, and calculation is completed to obtain a reconstructed image.

The invention discloses a pedestrian re-identification method based on multi-characteristic integration; the method comprises the following steps: 1, collecting multi-characteristic information; 2, carrying out tolerance model learning on the multi-characteristic information collected by step1, and building a plurality of coupling models based on a single-characteristic; 3, weight learning: distributing a corresponding weight for each model according to the identification level presentation of the 4 models learned in step2 on a training data set; 4, multi-characteristic distribution: using each single-characteristic coupling model obtained in step2 to calculate an assigned image and a coupling image, thus obtaining the distance between images on each single-characteristic coupling model, ranking the distances to obtain a corresponding coupling result, carrying out weight integration for distances on each characteristic and weights obtained by step3 so as to solve the final combined distance, and further obtaining an optimized multi-characteristic coupling result.

The invention discloses a turbine pump inducer cavitation flow numerical prediction method based on low-temperature fluid, and belongs to the technical field of impeller machinery. The implementationmethod comprises the following steps: establishing an inducer model and an inlet and outlet runner model according to actual working conditions, and dividing grids; Establishing a computational fluidmechanics model, carrying out rotation correction on the turbulence model, and carrying out thermodynamic correction on the cavitation model; Setting and calculating boundary conditions of the single-phase steady flow to obtain a single-phase result; Calculating two-phase steady flow by taking a single-phase result as an initial value, and then changing inlet pressure and outlet mass flow to obtain calculation results of different working conditions; Calculating unsteady flow of the two phases by taking the two-phase steady result as an initial value to obtain an unsteady calculation result; By analyzing a calculation result, accurate flow field characteristic distribution, inducer performance and radial force dynamic characteristics of the inducer under the action of a flow field are obtained, the inducer is assisted to be designed and optimized, and the experiment cost and time are saved.

The invention discloses a corrosion-resistance optimizing treatment process for Cr-Ni type stainless steel and a Cr-Ni type corrosion-resistant plate based on the same, belonging to the technical field of metallurgy. The optimizing treatment process comprises the steps of cold rolling and annealing, and is characterized in that after the plate is subjected to 3-5% of cold rolling, the deformed plate is subjected to multi-step annealing treatment. Common large-angle grain boundaries in the microscopic structure of the Cr-Ni type corrosion-resistant plate subjected to optimizing treatment have the preferable distribution characteristic on the {111} crystal face, and the proportion of the special grain boundaries is not lower than 75%; and the Cr-Ni type corrosion-resistant plate has favorable corrosion resistance. On the premise of optimizing the grain boundary characteristic distribution of the alloy, the invention shortens the annealing cycle, and certain induced common large-angle grain boundaries have the characteristic of low energy, thereby obviously enhancing the intergranular corrosion resistance of the alloy.

The invention relates to a diesel oil / natural gas dual-fuel engine emission control strategy which is characterized by comprising the following specific steps: S1, sensing the crankshaft position andjudging load; and S2, controlling an oil gas shared ejector. Ejection of diesel oil and natural gas can be controlled in real time through the oil gas shared ejector and a dual-fuel ECU to adjust concentration distribution and characteristic distribution of fuel in an oil cylinder in real time; and aiming at different working conditions and the combustion characteristics of fuel, and by combiningthe flexible ejection strategy and proper EGR, the mixing characteristic and the ignition characteristic of the interior of the cylinder can be effectively controlled to optimize cylinder interior combustion and reduce emission of NOX, PM, HC and CO at the same time.

The invention discloses a statistical modeling method for a radiofrequency variable capacitor. The statistical modeling method comprises the following steps: determining all model parameters needing to be extracted for device modeling; classifying the parameters; performing a scattering parameter test on a device, and establishing a reference model not including process fluctuation of the device; acquiring the fluctuation characteristic of the scattering parameter of the device on a whole wafer, and determining a fluctuation data part; constructing an index for performing statistical modeling on the device based on the scattering parameter of the fluctuation data part and a network parameter converted from the scattering parameter; calculating each element value in a sub-circuit model based on the index, performing sensitivity analysis and sequencing of the element values relative to an index value, and selecting an element with highest sensitivity; performing sensitivity analysis and sequencing of parameter values relative to the element values respectively specific to each type of parameter groups in an element extension formula; selecting a parameter with highest sensitivity, and fitting the statistical characteristic distribution of the device to obtain a statistical model of overall fluctuation distribution of the device. The modeling process is rapid and clear.

The present invention provides a light modulation element in which a visible image is written by simultaneously conducting irradiation of the light modulation element with exposure light according to image information which corresponds to the visible image and application of a voltage, having: a pair of electrodes to which the voltage is applied; a photoconductive layer which, when the light modulation element has been irradiated with the exposure light, shows an electric characteristic distribution corresponding to the intensity distribution of the exposure light; a liquid crystal layer to which a partial voltage derived from the voltage applied to the pair of electrodes and having a distribution corresponding to the electric characteristic distribution of the photoconductive layer is applied to record a visible image having an optical characteristic distribution corresponding to the distribution of the partial voltage; and a light shielding layer disposed between the photoconductive layer and the liquid crystal layer, wherein the photoconductive layer, the liquid crystal layer and the light shielding layer are disposed between the electrodes, and the light shielding layer contains a resin including partially saponified polyvinyl alcohol; and an image display device including the same.

The invention relates to the technical field of traffic safety, and specifically discloses a highway lane changing risk evaluation method based on transverse and longitudinal kinematic characteristicdistribution. The method comprises the steps of: calculating a maximum value of transverse acceleration as a driving criterion of a vehicle based on transverse kinematic characteristic distribution; calculating a driving criterion of the vehicle based on longitudinal kinematic characteristic distribution; calculating a comprehensive driving criterion; calculating a real-time minimum safe distance;calculating a transverse risk index; calculating a longitudinal risk index; and performing risk detection to obtain a result. According to the highway lane changing risk detection method based on transverse and longitudinal kinematic characteristic distribution, a personalized driving discrimination method capable of representing the personalized characteristics of drivers and a lane changing risk evaluation method are established based on a highway lane changing process; a reference is provided for the optimization of lane changing auxiliary; the intelligent process of vehicles is accelerated; the probability of traffic accident occurrence and the casualties and property loss caused by traffic accidents are reduced.

The invention provides a semantic marking method for an image scene based on geodesic transmission, which comprises the steps of: obtaining a rough semantic probability graph corresponding to an image scene by an object judging method; estimating the color characteristic distribution and the boundary characteristic distribution of the image scene; defining a geodesic distance on a mixed flow pattern by combining the color characteristic distribution with the boundary characteristic distribution; determining a point set with maximal local probability by a mean-shift algorithm as an initial seed point of geodesic transmission for various semantic categories; and determining the shortest geodesic distance of each point in the scene by a quick transmission algorithm based on a priority sequence for the geodesic distances of the defined various semantic categories so as to obtain the accurate semantic mark of the image scene. The invention can be extensively applied in semantic informationmarking of computer vision systems in fields of military, aviation, aerospace, monitor and manufacture, and the like.

An optical tomographic measuring device that includes: an illuminating component, a plurality of light-receiving components, a storage component, a specifying component, an acquiring component, and a constructing component is provided. The specifying component specifies a position of the measurement plane in the body length direction. The acquiring component acquires, from the storage component, an optical characteristic distribution that corresponds to the position specified by the specifying component. The constructing component constructs a density distribution of fluorescence in the measurement plane, on the basis of intensities of the fluorescence received at the respective light-receiving components and the optical characteristic distribution acquired by the acquiring component.

Provided is a solid state image pickup element which can exponentially reduce the in-plane photoelectric conversion portion characteristic distribution that is created in forming color filters by a common photolithography technique and which, when color filters are formed by split exposure, can reduce image non-uniformity between exposure regions in a picked-up image, and a method of manufacturing the same. The method includes: applying negative type color resist for forming first color filters onto an entire surface of a given film; forming the first color filters by irradiation of given portions with exposure light and by subsequent development; applying negative type color resist for forming second color filters onto the entire surface of the first color filters while covering the first color filters; and forming the second color filters by irradiating an area smaller than a region that is surrounded by the first color filters with exposure light and by subsequent development. Further, in the solid state image pickup element with color filters of plural colors, one of adjacent color filters is tapered at the edges while the other color filter is reversely tapered at the edges, allowing the adjacent color filters to fit to each other. Furthermore, adjacent color filters are wedged at the edges, allowing the adjacent color filters to fit to each other.

The invention provides a three-dimensional module comparison and search method based on nuclear density estimation, which comprises the steps of normalization of three-dimensional model coordinate data, extraction of three-dimensional module characteristics, resampling of the characteristics, nuclear density estimation and three-dimensional model comparison. The normalization of the three-dimensional model coordinate data comprises gridding subdivision, translation normalization of a three-dimensional model and scaling normalization of the three-dimensional model. The resampling of the characteristics is performed by merging adjacent characteristic pairs to reduce the amount of the characteristic pairs from n (n-1) to n. The nuclear density estimation is realized by performing multidimensional density estimation to generate a characteristic distribution function of a three-dimensional model. The three-dimensional model comparison is realized by comparing the similarity of the corresponding characteristic distribution functions of three-dimensional modules by utilizing KL distance to compare and search the three-dimensional models. The nuclear density estimation has greater flexibility and generality for characteristic distribution modeling of three-dimensional models with different types and different shapes. Multidimensional nuclear density estimation can utilize abundant multidimensional shape characteristics, and can better depict the characteristics of the models compared with a method in which data with different dimensionalities are simply combined.

The invention discloses an improved electric vehicle charging non-linear load harmonic wave electric energy metering method. On the basis of the Fourier transform algorithm, the method is performed as follows: using a time-scale analysis method to construct a data analysis algorithm which is consistent with the charging characteristic distribution of an electric vehicle; checking the localized characteristics of a non-stable signal; adjusting the widths of the time window and the frequency window to realize the variable time-frequency resolution; optimizing the electric vehicle non-linear load harmonic wave electric energy metering method; identifying and separating the steady-state harmonic waves and non-steady-state harmonic waves; conducting time-frequency decomposition, stratification and reconstruction to the signal; and achieving the purpose of accurate metering in the entire process. The method of the invention can separate the steady-state and non-steady-state harmonic waves generated by the nonlinear load during the charging operation of the electric vehicle, and performs the multi-scale refinement analysis of the electric power harmonic signal by the telescopic and translational operation of the time scale, which solves the insufficiency problem in Fourier transform in electric vehicle charging non-linear load harmonic wave electric energy metering, therefore, raising the accuracy of electric vehicle charging energy metering.