65 results about "Variance function" patented technology

The invention discloses a Gaussian process regression-based method for predicting state of health (SOH) of lithium batteries, relates to a method for predicting the SOH of the lithium batteries, belongs to the fields of electrochemistry and analytic chemistry and aims at the problem that the traditional lithium batteries are bad in health condition prediction adaptability. The method provided by the invention is realized according to the following steps of: I. drawing a relation curve of the SOH of a lithium battery and a charge-discharge period; II, selecting a covariance function according to a degenerated curve with a regeneration phenomenon and a constraint condition; III, carrying out iteration according to a conjugate gradient method, then determining the optimal value of a hyper-parameter and bringing initial value thereof into prior distribution; IV, obtaining posterior distribution according to the prior part; V, obtaining the mean value and variance of predicted output f' without Gaussian white noise; and VI, together bringing the practically predicted SOH of the battery and the predicted SOH obtained in the step V into training data y to obtain the f', then determining the prediction confidence interval and predicting the SOH of the lithium battery. The method provided by the invention is used for detecting lithium batteries.

The invention discloses a spatial variability layout plan generation method of heavy metal content in reclamation land soil, belonging to the technical field of environmental monitoring. The spatial variability layout plan generation method comprises: obtaining the soil sample of a sampling point in a reclamation land to be measured; carrying out data pretreatment to the numerical value of heavy metal content in the soil sample of a sampling point to respectively obtain the quasi-interpolation subsample of each soil heavy metal element; carrying out exploring spatial data analysis treatment to the quasi-interpolation subsample to obtain each anisotropy parameter, step length and step number of the soil sample of the sampling point; carrying out function fitting to the numerical value of heavy metal content in the soil sample of the sampling point to obtain a theoretical semi-variogram model suitable for obtaining heavy metal content interpolation; and drawing the spatial variability predicting result layout plan. The layout plan obtained by the invention can accurately evaluate the spatial condition of the heavy metal content in the reclamation land soil.

The invention relates to a method for generating a comprehensive evaluation map of heavy metal pollution of polder soil. The method comprises the following steps: acquiring a sampling site soil sample of a to-be-detected polder; performing data pretreatment on the numerical values of the contents of heavy metal elements in soil, thereby respectively obtaining a quasi-interpolation sub-sample of each heavy metal element; performing exploratory spatial data analysis treatment on the quasi-interpolation sub-samples, thereby obtaining the anisotropic parameters, a step length and a step number of the sampling site soil sample; performing function fitting on the numerical value of the contents of the heavy metal elements in the sampling site soil sample, thereby obtaining a theoretical semivariance function model suitable for interpolation of the contents of the heavy metal elements; drawing a spatial variation forecasting result distribution map by utilizing the sampling site comprising outlier; and comprehensively evaluating the heavy metal pollution of soil and generating the comprehensive evaluation map of heavy metal pollution of soil according to relative standard. Compared with the prior art, the method provided by the invention has the advantages that the advantages of the predication and evaluation of the heavy metal pollution of the polder soil, such as accuracy and precision, can be improved.

The invention discloses a mineralizing abnormal extraction method based on a multifractal Krige method. The mineralizing abnormal extraction method is characterized by comprising the following steps of: performing the inspection of normal distribution on an exploration geochemical data set to be analyzed, calculating an experimental variance function value for the data set which submits to the normal distribution, and fitting the experimental variance function value by a theoretical variance function model to acquire relevant parameters in the theoretical variance function model; and solving an abnormal lower limit value effectively according to a multifractal Krige interpolation result on the basis of establishing local high-value information and highlighting weak anomaly to acquire a reliable mineralizing abnormal area for prospecting and exploration.

The invention provides a method for evaluating the influence of speckle coherence on the ranging accuracy of a single-photon laser radar. The method comprises the steps of setting the parameters of the single-photon laser radar system, calculating the autocorrelation function of a receiving aperture of the single-photon laser radar system and the normalized covariance function on the intensity ofthe receiving aperture to calculate the speckle degree of freedom M; calculating the average signal photon number Ns according to the laser radar equation, and calculating the total noise rate Nn of the laser radar system; differentiating time through the detection probability based on the root mean square pulse width [Sigma]s of the laser pulse to obtain the detection probability density functionfs(t) of the echo signal with respect to time t, and obtaining the mean value shown in the description and the variance Var of the time when the detector detects the target point; and obtaining the influence of the speckle coherence on the ranging accuracy of the single-photon laser radar according to the drift error Ra and the random error Rp. The invention has good compatibility, can provide guidance for the system parameter design of the laser radar, improve the detection probability and reduce the ranging error as much as possible under the restraint of satisfying the false alarm probability.

This invention discloses an illumination light source device for adaptively adjusting light intensity composed of an image card, a computer, PWM control circuit, LED light source and a light source main circuit, among which, the data output end of the image card is connected with the input end of the computer its output end is linked to the input end of PWM control circuit, its output end is connected with the control end of the light source main circuit, output of which is linked to the input of LED source, its operation module is formed by an image quality quantization process unit computing image gray standard difference and variance functions according to signals input by the card and maximum unit of the optimum supply current corresponding value.

The invention provides a method and a system for estimating ground PM2.5 based on a space-time regression Kriging model. The method comprises the steps of re-sampling ground PM2.5 observation data of a to-be-estimated region to a created mesh, and performing matching, wherein the matching process comprises the steps of averaging the ground PM2.5 observation data monitored in the same day by all PM2.5 stations in a mesh unit corresponding to the to-be-estimated region in the created mesh, and then assigning the averaged data to the corresponding mesh unit; calculating an experimental variance function of a residual error according to the ground PM2.5 observation data of the matched to-be-estimated region, and determining a space-time variance function model according to the experimental variance function of the residual error; performing fitting on the space-time variance function model by adopting a least square method; and estimating a ground PM2.5 concentration value of the to-be-estimated region by adopting the space-time regression Kriging model according to a fitting result of the space-time variance function model. Through the method and the system, the PM2.5 estimation precision can be improved.

The invention discloses a data scientific visualization-oriented inverse distance weighting mixed interpolation method. For advantages and disadvantages of a conventional interpolation method, a new mixed interpolation method is proposed in combination with a parameter estimation point azimuth search method and an improved interpolation point cyclic interpolation mode by mixing an inverse distance weighting interpolation method with a Kriging interpolation method. The data interpolation method comprises the steps of (1), obtaining required data information; (2), calculating an experimental variance function; (3), selecting a proper variance function theoretical model for performing parameter fitting, and determining variance function parameters; (4), determining a participation point search scheme; (5), determining the total number of required interpolation grids and performing interpolation frequency division; and (6), performing cyclic interpolation until an interpolation effect is achieved.

The invention relates to an automatic focusing searching algorithm, which combines a Hill Climbing searching method and a function approaching method. In the algorithm, the Hill Climbing searching method adopts a two-stage algorithm combining coarse and fine focusing: in the coarse focusing, a long pace selects a gradation variance function to fast approach to a focusing position considering the rapidity of the algorithm; and in the fine focusing, a shorter pace adopts a Laplacian function to accurately focus the position considering the sensitivity of the algorithm. A focusing interval is narrowed by comparing three images and an optimal focusing position is fitted using the function approaching method in small intervals. By using the method, image acquisition and assessment times are reduced, an automatic focusing duration of a system is shortened and the searching efficiency of the algorithm is improved; and the influence of a local peak can also be avoided by comparing evaluation function value of the three continuous images. An extreme point is provided by fitting analysis, which makes the extreme point more close to the position of the optimal focusing point, thus algorithm precision is improved significantly.

The invention discloses a forest soil nutrient spatial prediction method based on artificial neural network Kriging interpolation, which comprises the following steps of: obtaining environmental factor grid data; calculating to obtain a forest soil nutrient spatial distribution diagram based on a multi-layer perceptron neural network; carrying out residual calculation between a measured nutrient value and a predicted value; carrying out analysis and verification on the prediction residual of the neural network; carrying out semi-variance calculation of the residual simulating a model determined by a semi-variance function to obtain model types and parameters; carrying out ordinary Kriging interpolation on the parameters of semi-variance model parameters to obtain the spatial distribution of the neural network prediction residual; adding a forest soil nutrient grid and a prediction residual grid based on a multi-layer perceptron neural network to obtain the forest soil nutrient spatialdistribution diagram based on the artificial neural network Kriging interpolation. The prediction precision of the method is obviously improved compared with a method by using only a multi-layer perceptron neural network model or ordinary Kriging interpolation.

The invention discloses a geostatistics-based wind power station wind speed spatio-temporal data modeling method. The method comprises the following steps: 1, constructing a space structure matrix of a wind power station according to geographical location information of fans, space convariance function of wind power station wind speed and variation function of wind power station wind speed to represent the spatial correlation of input wind speed of each fan; 2, performing layered modeling on the input wind speed of each fan by a universal Kriging method and a Bayesian algorithm and estimating model parameters by adopting Gibbs sampling; and 3, predicting the forward P steps of the wind speed of each fan to acquire simulation samples of P step forward prediction distribution of the wind speed of each fan, and sampling and averaging to acquire the optimal prediction result of the wind speed of each fan, wherein P is more than or equal to 1. By the method, the time and space correlation of wind speed data of different fan positions is comprehensively analyzed based on the physical characteristics of wind, so that a more accurate prediction model is built for the whole wind power station, and the prediction result is better than that of the traditional method.

The embodiment of the invention discloses an image tampering detection method and device. The method comprises the following steps: segmenting a to-be-detected image into multiple sub-images with overlapping regions by using a sliding window; extracting feature vectors of various sub-images by using a discrete analytic Fourier-Merlin transform method; performing multiple projections on various feature vectors by using a local sensitivity Hash algorithm to respectively multiple projection values with the same number of each sub-image; judging whether a target sub-image pair meeting a preset projection condition is existent; regarding that the to-be-detected image is not tampered if the target sub-image pair meeting the preset projection condition is nonexistent; computing a filter referencevalue of each target sub-image in the target sub-image pair by using a Weber local description sub-method and a variance function if the target sub-image pair meeting the preset projection conditionis existent, and judging whether each filter reference value is greater than a threshold; regarding that the to-be-detected image is tampered if the filter reference value is greater than the threshold; or regarding that the to-be-detected image is not tampered if the filter reference value is not greater than the threshold. By using the technical scheme provided by the application, the accuracy of image tampering detection is improved, the computing complexity is reduced, and the recall ratio is improved.

The invention discloses a quality control method of a ground temperature observation material. The method includes the following steps: firstly acquiring temperature data of a meteorological station within an object region, then resolving an experimental variance function value, and using an arctan function that is proposed to carry out fitting on the experimental variance function value, then resolving the weight of each neighboring meteorological station to an object meteorological station by using a Kriging equation set, further using gradient Kriging method to estimate the temperature values of respective meteorological station and evaluate the precision of the temperature values, and finally conduct threshold examination on a temperature estimation value and an original observation value so as to determine whether the data is right or not and correct the data accordingly. According to the invention, the method addresses the problems of insufficient consideration of impact of space distribution, temperature space relativity and variance thereof and landform on temperature of current quality control methods.

The canopy space statistics quantitative estimation for remote image with high spatial resolution, canopy parameter extraction image testing half variance function aeolotropy analysis, computing the half variance function value in four different directions, judging the significance of the variance through special software, deciding the canopy width value in three conditions, such as the unobvious, obvious and obvious but only demanding the average canopy range. Solving higher canopy density, it can not tell the canopy range estimate directly from the image.

A system for large scale data modelling is described. The system includes at least one data measurement sensor (230) for generating measured data, a training processor (240) to determine optimized hyperparameter values in relation to a Gaussian process covariance function including a sparse covariance function that is smooth and diminishes to zero outside of a characteristic hyperparameter length. An evaluation processor (260) determines model data from the Gaussian process covariance function with optimised hyperparameter values and measured data. Also described is methods for modelling date, including a method using a Gaussian process including a sparse covariance function that diminishes to zero outside of a characteristic length, wherein the characteristic length is determined from the data to be modelled.

The invention relates to the field of nonlinear time-varying system optimization control, in particular to a control strategy for achieving multivariable PID in a PLS frame on the basis of a Gaussianprocess model. The problems that by means of existing control strategies, interaction among multiple loops cannot be eliminated, a built model is not complete, and the control performance is poor aresolved. The control strategy for achieving multivariable PID in the PLS frame on the basis of the Gaussian process model comprises the steps that 1, parameters of a PID controller are given; 2, the GP(Gaussian process) model is used for achieving the control strategy of multivariable PID, wherein firstly, an MIMO system is subjected to decoupling, secondly, the model uncertainty is improved, thirdly, a covariance function is chosen, and the parameters are optimized; 3, the PID controller is set, wherein a gradient-based optimization algorithm is used for using the GP model for adjusting the PID controller. The control strategy for achieving multivariable PID in the PLS frame on the basis of the Gaussian process model has the advantages that the GP model is used for providing a predictionvariance, and the prediction reliability of a local random area is shown; cross coupling effects brought by a multivariable control process are taken into account, through the PLS frame, the MIMO system is decoupled into single loops, and then based on the GP model, the PID controller parameters are adjusted separately.

The invention relates to a robustness sorting learning method based on multi-objective particle swarm optimization and an application thereof, and the method comprises the following steps of 1, designing an effective deviation function and a robustness variance function of a sorting model based on a deviation-variance equalization theory, and constructing two optimization performance indexes of sorting learning; step 2, on the basis of a multi-objective particle swarm optimization algorithm framework, iteratively optimizing the two objectives of the effectiveness deviation function and the robustness variance function of the sorting model on the sorting learning data set to train the sorting model so as to generate a sorting model archiving solution set; and step 3, selecting a Pareto optimal sorting model with a maximum net flow sorting value from the sorting model filing solution set generated in the previous step as a trained final sorting model based on the idea of a preference sequence structure assessment method PROMEHEE II in a multi-attribute decision theory. Compared with the prior art, the method has the advantages of improving the overall user satisfaction, enhancing theuser experience and the like.

The invention belongs to the fine particle detection field, in particular to a fine particle prediction and traceability method based on the combination of Gaussian process regression and firefly algorithm. The method comprises the following steps of optimizing the firefly algorithm to obtain a firefly optimization algorithm; creating a sample training set of the model and selecting a covariance function for Gaussian process regression model; carrying out the parameter optimization on the equation of the Gaussian process regression model by using an improved particle swarm optimization algorithm; according to the optimized super-parameters, obtaining the model and outputting the prediction. The invention provides the fine particle prediction and traceability method based on the combinationof Gaussian process regression and firefly algorithm which combines the Gaussian process regression with the firefly algorithm, can not only predict the spatial distribution and short-term variationtrend of fine particles, but also trace back the multi-pollution sources dynamically, thereby accurately locating the multi-pollution sources of fine particles.

The invention discloses a bat algorithm and Otsu algorithm-based infrared image segmentation method. The method comprises the steps of S10: obtaining a to-be-segmented infrared image; S20: taking an interclass variance function in an Otsu algorithm as a target function of a bat algorithm, and taking a segmentation threshold variable corresponding to the interclass variance function as a bat position variable in the bat algorithm; and S30: based on the bat algorithm, calculating a corresponding bat position when the target function is maximal, and segmenting the to-be-segmented infrared image into a background region and a target region by taking the bat position as an optimal segmentation threshold. In addition, the invention discloses a corresponding system. The to-be-segmented infrared image can be segmented quickly. Compared with an existing infrared image threshold segmentation method, the threshold selection speed is effectively increased and a foundation is laid for real-time extraction and analysis of temperature field characteristics of subsequent equipment.

The invention discloses a medical image segmentation method and a device, wherein the method comprises: step S11) obtaining magnetic resonance angiography image; step S12) for the magnetic resonance angiography image, using the Otsu threshold method to divide it into an interested foreground area and a background area; and calculating difference value between the pixel mean gray value of the foreground area and the pixel mean gray value of the background area corresponding to the maximum of the variance function of the foreground area and the background area; and Step S13): determining the difference value between the gray mean values of the internal images and external images of the evolution curve of the image segmentation model C-V according to the difference value; and segmenting the magnetic resonance angiography image according to the determined difference value between the gray mean values of the internal images and external images of the evolution curve and obtaining a segmenting result. The medical image segmentation device includes an image obtaining module, a difference value determining module, and a segmenting module. The medical image segmentation method and the device of the present disclosure improve the segmenting effect and the processing speed, meeting the requirements.

The invention discloses a least-square collocation model of a satellite gravity gradient tensor diagonal three-component inversion earth gravity field and a modeling method, and belongs to the field of satellite gravity measurement. The method comprises the steps that starting from an orthogonality rule of a spherical harmonics function, Txx and Tyy components and Tzz, Txx and Tyy components are combined separately, practical calculation formulas of auto-covariance and cross covariance functions among the satellite gravity gradient tensor diagonal three-components and covariance functions of the components and gravitational potential coefficients are derived, and then the least-square collocation model of the Txx and Tyy two-component and Tzz, Txx and Tyy three-component inversion earth gravity fields is obtained. Accordingly, the theoretical basis is provided for fully utilizing the satellite gravity gradient tensor diagonal three-component inversion earth gravity field, and data resources of a gravity gradient measurement satellite GOCE launched by the ESA are fully utilized.

The invention relates to a method for automatically extracting phenology information of earth surface vegetation based on the fitting variance of a Gaussian function. A range which is not only a space concept but also a time concept is applied to operation, the range is calculated, the obtained range represents range values of the whole NDVI time sequence set, the range values represent a period in which the variance function changes most, and the change period represents the growth period of the surface land vegetation of a research area. The method comprises the steps of 1) preprocessing initial remote sensing images; 2) determining the fitting goodness; 3) determining the range reasonableness; and 4) completing extraction of the phenology information. The method can be used to effectively solve the problems including that the fitting parameters are too many, the calculation process is complex and manual determination is subjective, results are obtained via calculation based on computer in the whole course, subjectivity of manual determination is eliminated, and it is proved that result data is highly correlated to the practical phenology information of the earth surface vegetation in the research area and is quite indicative and representative.

The invention relates to a confocal microscope mode aberration correction method and belongs to the adaptive optics and confocal microscopy imaging technological field. The invention aims to solve theproblems of focal spot distortion and resolution reduction caused by aberration which is further caused by the assembly error of an existing optical system and the surface shape deviation of opticalelements. According to the method, the fourth order to eleventh order of Zernike term bias aberrations of different amplitudes are loaded sequentially; the values of the gray-scale variance functionsof corresponding images are calculated; corresponding aberration coefficients are calculated through using a centroid method; comparison is performed, if the aberration coefficients are smaller than 0.7rad, zero clearing is performed, and other terms are pre-corrected; aberration coefficients are re-calculated through linear computation; and aberration correction is performed through using the sums of the coefficients obtained through the above two different times. According to the confocal microscope mode aberration correction method of the invention, wavefront distortion caused by the assembly error of the optical system and the surface shape deviation of the optical elements is compensated inversely through a spatial light modulator, and therefore, the aberration influence of a confocalmicroscope system can be effectively eliminated, and the imaging quality and resolution of the confocal microscope system can be enhanced.

The invention provides a focusing method based on a quantum particle swarm optimization algorithm for focusing an image during image shooting. The method comprises the following steps: step 1: using aplurality of particles located in a three-dimensional space to respectively represent gray values of pixel points in the image, and randomly setting the gray values represented by the particles; step2, performing average distribution on the initial positions of the particles in a quantum particle swarm according to a position formula; step 3, calculating fitness values of the particles by usingan average gray value variance function of a foreground image and a background image of the image, and obtaining an optimal segmentation threshold by using the quantum particle swarm optimization algorithm in combination with a domain search method; step 4, segmenting the image into the foreground image and the background image according to the optimal segmentation threshold; step 5, selecting a focusing area according to the center of gravity of the gray value of the foreground image; and step 6, using a gray level difference method as an image definition evaluation function, and determiningthe position of a lens according to a function value calculated by the image definition evaluation function so as to complete the focusing.

The invention discloses a complex surface self-adaptive sampling method. The content of the method comprises the steps that a sampling quantity is determined according to sampling precision; complex surface projective points are included into a given square region and converted into two-dimensional projective points with density properties; sample points are randomly generated under set initial temperature; search is performed on the sample points, and a mean value of two-generation generators is calculated and made to approach a centroid point step by step; a variance function is used as a criterion to determine the goodness / badness of solutions, and for deteriorative solutions, a random probability is set to expand a solution space; the generators are updated, and the coincident precision of the generators and the centroid point is judged; the generator with the highest coincident precision is selected to serve as the centroid point and a globally optimal solution, and return is performed; and the optimal solution is mapped back to a complex surface to serve as an optimal sampling solution for output. The method is high in calculation speed and efficiency, and measurement efficiency and measurement precision of surface self-adaptive sampling are improved.

The invention discloses a method for determining the number of reasonable samples in forest aboveground biomass remote sensing estimation. The method comprises steps of taking two tree species, i.e.,pinus alpine and fir yunnanensis, as research objects, combining a Landsat 8 image and ground survey sample plot data, taking a k-NN remote sensing estimation model as a case and taking a root-mean-square error as an evaluation index based on a geostatistics semi-variance function theory; and calculating to obtain the number of samples when the precision of the pinus alpine and fir clovershanensismodels is optimal, and determining a reasonable sample number range. In order to discuss uncertainty of the number of samples in quantitative remote sensing inversion, the uncertainty influence of the number of samples in forest aboveground biomass estimation is analyzed based on a geostatistics semi-variation function theory and a k-NN model, and a reference basis is provided for a problem thatthe number of samples in traditional quantitative remote sensing inversion is difficult to determine.

The invention relates to a method and system for determining alarm duration statistical distribution stability. The method includes the steps of sorting acquired original signals to obtain an alarm duration set of all alarming signals, constructing an alarm duration probability density function and a standard variance function according to the alarm duration set, removing the alarm duration of low occurrence frequency from the alarm duration set, obtaining the screened alarm duration set, plugging the values of the screened alarm duration set respectively in the alarm duration probability density function and the standard variance function, and determining the statistical distribution stability of the alarm duration according to the ratio of the alarm duration probability density function to the standard variance function. Whether the alarm duration in historical data has regularity on statistical sense can be determined, an alarm generation mechanism is optimally designed according to stable alarm duration statistical distribution, and interference alarm existing in an alarm system can be reduced.

The invention discloses an assistant decision-making method for gas sensor adjustment on a working face. The method comprises the following steps: assuming a moment before gas sensor adjustment is t,extracting q data at a moment t-1 and before the moment t-1 from a gas concentration data sequence monitored before the gas sensor adjustment according to a time sequence to form basic data C; performing training according to a Gaussian process regression theory to obtain a Gaussian regression model GP(C,y); taking gas concentration monitoring data at the moment t before the gas sensor adjustmentas test input; obtaining a mean value function and a variance function of gas concentration normal monitoring data corresponding to the moment t with the trained Gaussian regression model; acquiring aconfidence interval according to the mean value function and the variance function of the gas concentration normal monitoring data corresponding to the moment t in terms of the characteristic of Gaussian distribution; and comparing a gas concentration actual monitoring value after the gas sensor adjustment with the obtained confidence interval to judge the effectiveness of the gas sensor adjustment.

The software aims to automatically extract a newly-added construction land in a high-resolution SAR image and belongs to the field of image recognition. Based on a variation function method, the software can automatically and accurately extract the newly-added construction land in the high-resolution SAR image, and mainly comprises the following steps of: first, preprocessing (filtering) an input image; second, calculating a variance function to obtain the texture features of two images; third, calculating a ratio of the texture features of the two images to create a difference image; fourth, subjecting a ratio image to threshold segmentation to generate a binary image that is a preliminary newly-added construction land extraction result; and fifth, post-processing the binary image to generate a final extraction result and to perform accuracy evaluation. The extraction accuracy of the software can reach more than 80% according to a built-in test, and the soft ware can be applied to the detection of building change areas such as urban planning, determination for demolition compensation, and the like.

The invention discloses a multi-target threshold image segmentation method for fuzzy information and statistical information in a fusion interval. The method comprises the steps of: inputting an image to be segmented and converting the image into a grayscale image; setting the initial population number of the image as N, the maximum number of iterations as G and the maximum threshold number as Smax, and then dividing the population into several grouping population Qs of the same size according to the threshold number; conducting multi-target evolution for the obtained grouping population Qs through simultaneous optimization of the interval modulus entropy function and the inter-class variance function based on the linear intercept histogram to allow each grouping population to obtain a group of non-dominated solution sets, selecting an optimal solution in the non-dominated solution set of each grouping population through weighted ratio of the inter-class variances, the optimal solution being the optimal threshold number and the optimal threshold value; and conducting mark assignment for the pixel points in the original image according to the optimal solutions, and obtaining the final segmentation results. The method can realize the adaptive threshold image segmentation, and a satisfactory result can be obtained for a noisy image.