40 results about "Chi-squared distribution" patented technology

The invention discloses a new method for RAIM (receiver autonomous integrity monitoring) based on a satellite selecting algorithm in a multimode satellite navigation system. The method comprises the steps of first determining space position information of satellites according to a navigation message and eliminating satellites with a small elevation angle according to a shielding angle; determining an observation matrix including only one clock correction item according to clock correction conversion factors in the navigation message; selecting p satellites from N visible satellites so as to be used for positioning calculation of a receiver, acquiring a satellite combination, which enables the GDOP (geometric dilution of precision) to be minimum, through the satellite selecting algorithm to act as calculating satellites, and determining a weight matrix in WLS (weighted least squares) according to parameters such as the carrier-to-noise ratio, the loop bandwidth, pre-check integral time and the like of satellite signals; carrying out RAIM availability detection according to a false alarm rate and a missed alarm rate which are preset by the receiver, and calculating a pseudo-range residual error threshold value after positioning according to the false alarm rate and a degree of freedom in Chi-squared distribution; carrying out global detection at first, then carrying out local monitoring in a circumstance that a fault satellite exists, determining calculation satellites again through satellite selection, and finally carrying out positioning calculation through selecting satellite combinations within the threshold value. The method disclosed by the invention is simple, high in fault recognition rate, not only applicable to multi-mode and multi-fault satellite navigation systems, but also applicable to single-mode and multi-fault satellite navigation systems, thereby providing new ideas for carrying out RAIM by a modern GNSS (global navigation satellite system).

The invention discloses a track correlation method by taking topological sequences as characteristics, which comprises steps of constructing target topological sequences of a single sensor and matching multi-sensor topological sequences and belongs to the field of track correlation of multi-sensor networking. The method comprises the following steps of: forming the topological sequences of each target on the basis of a target position data set of the single sensor; filtering the topological sequences of the each sensor according to the similarity of a spatial location to construct F statistic which is submitted to chi square distribution; and judging that the topological sequences which meet statistical tests are from the same target by angle fine turning. In the track correlation method by taking the topological sequences as the characteristics, the topological sequences are taken as a state variable for identifying the target and the method has high accuracy, higher robustness, and high correlation performance.

The invention provides a data dynamic labeling method and device for coarse-grained text classification. The method comprises the steps of balancing the labeling data according to the label category proportion; constructing a text TF-IDF word frequency matrix; performing feature screening by using chi-square distribution to obtain a training data set; training the training data set by using a machine learning algorithm to obtain an initial model; acquiring a test data set, and labeling data of a first preset data volume in the test data set by using the initial model to obtain predicted labeling data; obtaining data corresponding to each data annotation label obtained by respectively extracting a second preset number of pieces of predicted annotation data according to the label category for auditing; adding the data corresponding to each data labeling label into a training data set, and training the training data set by using a machine learning algorithm to obtain a correction model; judging whether the training data volume meets a second preset data volume or not, and if not, continuing to execute the process; wherein the storage correction model is a prediction model.

Disclosed are a method and apparatus of adaptive transmission signal detection based on a signal-to-noise ratio and a Chi-squared distribution for decoding a reception signal in the form of multi-streams into a transmission signal, the method comprising: receiving a reception signal vector; estimating a channel matrix of a channel that the reception signal vector experienced; performing QR decomposition based on the estimated channel matrix; calculating dimensions of divisional spaces adaptively on the basis of a signal-to-noise ratio (SNR) or a signal-to-interference noise ratio (SINR); detecting a maximum likelihood point having a maximum likelihood value among lattice points within each divisional space optained on the basis of results from performing the QR decomposition and calculating dimension of the divisional spaces; and recovering a transmission signal vector based on the detected maximum likelihood points.

The invention discloses a high-resolution remote-sensing image change detection method facing targets and based on an integrating strategy. A high-resolution optical remote-sensing image registered through two time phases is input, and multi-resolution cutting is carried out on the input high-resolution image; change detecting is carried out on divided units (targets), and the mean value and the variance of each target are calculated; the mean values and the variances of all the targets are combined to form a new vector, changing regions and non-changing regions of the image are fixed according to chi-squared distribution; detection results are integrated on the basis of a voting method, and accordingly a change detection result is more stable and more accurate.

The application discloses a keyboard input abnormality detection method. The method comprises: obtaining an operation of a user for a keyboard in a mobile terminal; generating Z values corresponding to a plurality of dimensions according to the operation; generating a chi-square value corresponding to the operation according to the Z values corresponding to the dimensions; if the chi-square value is greater than a preset threshold, determining that the operation is an abnormal operation; and if the chi-square value is less than or equal to the preset threshold, determining that the operation is a normal operation. According to the keyboard input abnormality detection method provided by an embodiment of the application, based on normal distribution and chi square distribution principles, a keyboard input behavior is compared with historical data to determine whether a current input behavior is input by an account owner or not, so that whether a current account information input operation is an operation of the account owner or not can be identified more accurately, the identification rate is increased, and the misjudgment rate is reduced. The application furthermore discloses a keyboard input abnormality detection apparatus as well as a security prompt method and a security prompt apparatus.

The invention discloses a signal detection algorithm under a low signal-to-noise ratio condition. The detection algorithm comprises the steps that: (1), a random resonance system is used as the front end of an energy detector; random resonance is generated when an input signal y(t) and a resonance noise signal n<0>(t) pass through the random resonance system; the energy of resonance noise is converted into user signal energy; and a signal r<SR>(t) is output; (2), according to the output signal r<SR>(t) obtained in the step (1), the detection output statistical amount T(r)<SR> of the random resonance system is calculated; and (3), wherein both the detection statistical amounts T(r)<SR_ED> under the conditions of H<0> and H<1> obey noncentral chi-square distribution; if N is high enough, the two approximately obey normal distribution; and the false alarm probability and the correct detection probability can be obtained.

The invention discloses a method and a system applied to the in-situ crack detection of a wind generator blade. The method comprises the following steps: obtaining static responses of the wind generator blade under the condition that different loads are loaded by using an optic fiber grating sensor; calculating a Chi-square distribution curve among the static responses of the wind generator blade when the different static loads are loaded, and carrying multinomial reconstitution on the Chi-square distribution curve to obtain reconstruction curves; and by taking the Chi-square distribution reconstruction curves as different information sources, carrying out information fusion by using the reconstruction optimal accumulation theory to obtain an information fusion degree curve, and identifying the position of a crack of the wind generator blade according to the position of a maximum value of the information fusion degree curve. According to the method and the device, only the static responses of the wind generator blade when the different static loads are loaded are required; any data under the condition that the wind generator blade is complete is not required; the calculation is quick and accurate; the method and the device are simple, and liable to implement and are applicable to real-time health monitoring of the wind generator blade.

The present invention provides an abnormality detection optimization method oriented to power grid spatio-temporal data. The method comprises the following steps: (1) performing mesh division on a power distribution device according to spatial positions, then carrying out uniform division on power distribution power curves of acquisition devices in all meshes according to time, and defining a Gaussian distribution for power-distribution average power in each division; (2) establishing a null hypothesis and an alternative hypothesis; (3) estimating parameters, i.e. estimating values of testing parameters TP in the null hypothesis and the alternative hypothesis according to a maximum likelihood estimation method; (4) carrying out pruning optimization; (5) calculating a likelihood ratio of three-dimensional spacial data, wherein the higher the ratio is, the more obvious the abnormality of an area is; and (6) outputting first K abnormal areas by adopting a heap sort algorithm, and according to chi square distribution and in combination with a confidence level, acquiring an abnormality threshold. According to the abnormality detection optimization method provided by the present invention, power distribution network monitored data is classified so as to improve a fault detection technology aiming at a power distribution network and efficiently process interference of outside factors to power scheduling.

The invention relates to a determination method and system for a wind power state number upper limit in a wind power generating algorithm based on MCMC. According to the method, first, historical wind power exerting data is collected and sorted out, the data is divided into exerting states in different numbers, a state transition frequency number and a state transition probability matrix of all wind power exerting data points under the exerting states in different numbers are calculated respectively, a chi-square test is performed on the exerting states to determine a critical exerting state number conforming to Markov transition characteristics, and the critical exerting state number serves as a dividable exerting state number upper limit. A theoretical basis is laid for profound research of wind power exerting time sequence modeling. Through the chi-square distribution test method, a maximum wind power state number needing to be divided can be determined, and the precision of a wind power sequence generated through the MCMC method is improved to the maximum extent.

The invention discloses an information fusion method and system of a multi-sensor system, a computer device and a medium. A specific embodiment of the method comprises the steps of dividing a signal time length into a plurality of time windows, and obtaining a target sensor of a current time window; dividing the current time window into at least two sub-windows, judging whether the target sensor has a fault in the current time window or not according to the sum of the observation residual statistical property values of the target sensor at all moments in the current time window and a window preset chi-square distribution threshold; if yes, according to the time sequence of the sub-windows, judging the sub-windows where the faults of the preset chi-square distribution threshold values of the corresponding sub-windows are located sequentially according to the sum values in the sub-windows and the time sequence of the sub-windows, and judging the time when the faults occur moment by moment according to the characteristic values of all the moments in the sub-windows and the time preset chi-square distribution threshold values; and obtaining an information fusion result of the target sensor at the end moment of the current time window according to the fault condition of the target sensor in the current time window.

The embodiment of the invention provides an information fusion method and device of a multi-sensor system, and relates to the field of sensor information processing. The method can greatly improve adaptability of an information fusion algorithm and algorithm accuracy in complex cases, and includes: establishing an observation equation according to a preset random fault rate; determining target sensors which are in the multi-sensor system and can carry out information fusion at a current time; calculating theoretical values of observation values of the target sensors at the current time according to the observation equation; calculating values of the observation residual counting characteristics of the target sensors at the current time; according to size relationships of the values of theobservation residual counting characteristics of the target sensors at the current time and preset chi-square distribution values, judging whether the observation values of the target sensors at the current time have an error; and obtaining a result of information fusion of the target sensors at the current time according to whether the observation values of the target sensors at the current timehave the error.

The invention discloses a neighborhood contraction MRI de-noising method based on Chi-square unbiased risk estimation. The method comprises steps of: after estimating the noise standard deviation, squaring a noise image and then dividing the obtained value with the square of the noise standard variation so as to satisfy the property of the noncentral Chi-square distribution; then, carrying out stable Haar wavelet transform which has not been normalized so as to obtain high-frequency coefficient and low-frequency coefficient; deblurring the low-frequency coefficient by use of a bilateral filter; carrying out cyclic shift on the denoised wavelet coefficient on the high-efficient coefficient by use of the neighborhood contraction method based on the Chi-square unbiased risk estimation; and finally, averaging multiple shifted denoising images so as to obtain the denoised image.

The invention relates to a brightness enhancement method for two-photon microscopic images of low-brightness human tissues and cells, which comprises the following steps: low-brightness image data type is converted into double-precision type; the maximum values of the three color channels in the image are taken as the initial brightness evaluation maps of human tissue cell images; using correlation filter and linear optimization function, the accurate brightness evaluation map of human tissue and cell image is obtained; according to the principle of retinal imaging and cerebral cortex regulation, the exposure map of brightness enhancement is obtained; combined with chi-square distribution function optimization model and exposure mapping, a good brightness enhancement image is obtained. Theaim of that invention is to address the two-photon microscopic image of human tissue and cell with serious noise and complex background, not only can the visual information of the image be enhanced,but also higher fidelity of visual information and less distortion and distortion of brightness are obtained, thus solving the problem that more effective information can not be obtained in the two-photon microscopic image of human tissue and cell with low brightness.

Embodiments of the present invention provide a high-dimensional data similarity join query method and device based on mapping space partition. The method comprises: acquiring high-dimensional raw dataand mapping the raw data to one-dimensional space; determining a second distance threshold according to a first distance threshold and the chi-square distribution property, and dividing the one-dimensional space into a plurality of subspaces according to the second distance threshold; determining a number of the subspace corresponding to each raw data; obtaining a candidate data pair according tothe second distance threshold and the numbers of the subspaces; calculating an original distance of the candidate data pair and comparing the original distance with the first distance threshold to obtain a similarity query result. Device for performing a method is provided. As the high-dimensional raw data is mapped to the one-dimensional space, the raw data is divided in the one-dimensional space according to the second distance threshold, and then the similarity inquiry is carried out, and in this way, the computational complexity is lowered, the number of candidate results is reduced, andthe inquiry efficiency is improved.

The invention relates to a system for determining the activity of a mobile element (EM), including at least one motion sensor (CM) with at least one measurement axis, provided with an attachment means (MF) for rigidly connecting said motion sensor (CM) to said mobile element (EM), including: a filter (FILT) for selecting, for every measurement axis of the motion sensor, high frequencies above a first threshold (S1); a means (DET) for determining a one-dimensional high-frequency component (y(n)) equal to the square of the Euclidean norm of said high frequencies of the measurement axes taken into account by said motion sensor (CM); a means (CALC) for calculating, for every state, probability density functions (Py, i) of said high-frequency component (HF), said probability density function corresponding to each state according to a chi-square distribution of a degree of freedom equal to the number of measurement axes taken into account by the motion sensor (CM); and an analysis means suitable for defining states of the mobile element, said analysis means combines the probability density function of said high-frequency component for each state with the probability of changing between two consecutive states.

The invention provides a liver magnetic resonance R2* parameter graph drafting method. The method comprises steps that a liver magnetic resonance image is acquired, and a liver interest area is drafted; for each target voxel in the interest area, adaptive weight of neighborhood voxels can be acquired according to a vector similarity measurement model; in combination with attenuation signals and the adaptive weight of the neighborhood voxels, an R2* value of each target voxel is acquired according to a target optimization function model and a non-center chi square distribution noise first-order moment model, and the liver magnetic resonance R2* parameter graph is acquired. Through the method, the spatial resolution and the signal-to-noise ratio of the liver magnetic resonance R2* parameter graph can be effectively improved, and a more accurate liver iron-content distribution graph can be acquired.

The invention provides a senile dementia gene and site screening method based on improved Mendelian randomization, and belongs to the field of Alzheimer's disease related gene and site screening. Themethod comprises the specific steps: taking GWAS data as x, taking eQTL data as y, and taking SNP data as z; obtaining the effect bxy of x on y, the effect bzx of z on x, and the effect bzy of z on y;defining bxy: bxy = bze / bzx; calculating Z-score of SNP in an eQTL dataset; calculating Z-score of corresponding SNPs in a GWAS dataset; calculating a statistic Tsmr and obeying chi-square distribution with a degree of freedom of 1; performing the chi-square test of Tsmr and solution of P-value; and screening the values according to the sizes of the P-value. More genes and sites related to AD arediscovered, the risk of illness can be predicted during gene screening, and the method has extremely important significance for research and development of gene targeted drugs. Under the extreme condition, a gene knockout operation can be carried out to enable a patient to recover.

The embodiment of the invention discloses a micro-seismic event detection method and system based on chi-square distribution. The method comprises the following steps: S1, inputting a measured micro-seismic signal sequence S; and S2, judging a micro-seismic event according to the chi-square statistics gamma<K><2> of the micro-seismic signal sequence S. Specifically, a micro-seismic event is detected at a point K of the micro-seismic signal sequence S if the chi-square statistics gamma<K><2> is greater than or equal to gamma<alpha><2>, otherwise, no micro-seismic event is detected, wherein gamma<alpha><2> is a judgment threshold, and alpha is an F distribution parameter.

The invention provides a radius update method based on a sphere decoding algorithm and relates to the communication field. The method comprises steps that a present searching level, a radius initial value and respective radius adjustment values corresponding to other searching levels previous to the present searching level are acquired; based on non-center chi square distribution, an expectation value of a radius adjustment value sum corresponding to the present searching level is calculated; a radius of the present searching level is calculated according to the radius initial value, the respective radius adjustment values corresponding to other searching levels previous to the present searching level and an expectation value of a radius adjustment value corresponding to a non-searched level behind the present searching level. The method is advantaged in that the update radius acquired through solution is made to be more accurate, and thereby a problem of increase of calculation amount of the sphere decoding algorithm caused by amplification of the update radius is solved.

A method for making an impact response spectrum test profile includes the following steps of obtaining the impact data of which the impact response spectrum obeys the normal distribution, the energy spectrum density obeys the chi-square distribution and which belongs to the same population, through an impact test, and constructing the characteristic sample with the impact response spectrum of saidimpact data; 2 dividing the frequency bands of the characteristic sample along the frequency axis distribution into different frequency bands, and determining the flat spectrum of each frequency band; S3 calculating an upper tolerance limit of the flat spectrum, and generating a gauge spectrum according to the upper tolerance limit. The method of the invention has the advantages that the actual environment of the impact can be well reflected, and the result is more accurate and the like.

The invention relates to a ship instrument verification period determination method and system. Specific to the characteristic of large ship instrument verification amount, a Chi-square distribution model is established according to the instrument verification qualification situation of each batch; a judgment on whether or not an existing verification period is reasonable is made with a hypothesis testing method; and a corresponding adjusting method is provided. The verification period of instruments is adjusted reasonably, so that the technical problem of hidden danger to the accuracy and reliability of equipment performance caused by over-heavy verification task, cost increase and overlong verification period due to over-short verification period and high verification frequency in the prior art is solved. Meanwhile, the problems that a practical verification time interval is inconsistent with a specified verification period and that verification data is difficult to count and analyze since the instruments are under the influences of navigational duty and repair are solved. Through adoption of the model based on Chi-square distribution, specific reliability models of the instruments under verification do not need to be known, and certain universality is achieved.

The invention provides an industrial robot fault detection method, and belongs to the field of industrial robot process monitoring and fault detection. The problem that the fault detection accuracy ofan industrial robot is low is solved. According to the method, different Markov chains are set for a preprocessed data set to construct a dynamic potential reference model containing two types of hidden variables, and quality-related process variables and quality-independent process variable information can be extracted at the same time, that is, all process information serves as supervision items of model fault detection. The statistics can be calculated according to the distribution of the two hidden variables, the statistical thresholds and SPElim of the statistics T2 and SPE are calculated by chi-square distribution, the data operated in the working process of the industrial robot are collected, the statistics and SPEtest in the working process are obtained according to the trained dynamic potential reference model, and the statistics and SPEtest are compared with the statistical thresholds and SPElim, so whether the industrial robot breaks down in the working process can be judged. According to the method, the dynamic double-hidden-variable model is established, so the double hidden variables and the dynamics can be combined, the dynamic characteristics related and irrelevantto the quality in the process are fully extracted, and the accuracy of fault detection is improved.

The invention relates to an image noise level estimation method and device and a computer storage medium, and the method comprises the steps: obtaining an initial estimation value of the noise level of a noise-containing image, and obtaining an image flat block set of the noise-containing image according to the initial estimation value; according to the image flat block set, calculating an estimated value of the noise level of the noisy image based on chi-square distribution, and according to the estimated value, obtaining the image flat block set again and calculating the estimated value of the noise level of the noisy image, and stopping iteration until the estimated value of the noise level obtained by continuous two times of calculation is kept unchanged or iteration reaches a preset number of times, and obtaining a final estimated value of the noise level of the noisy image. By means of the method, on the basis of the chi-square distribution principle, the image flat blocks are repeatedly screened to conduct loop iteration to estimate the noise level, and the accuracy of noise level estimation of various noise images can be greatly improved.

The invention discloses a gene-environment interaction correlation analysis method based on a large sample, and the method comprises the steps: building a gene-environment detection linear hybrid model through collecting the gene sequence, environment information and phenotype information of a target individual, and solving the gene-environment detection linear hybrid model based on a preprocessing conjugate gradient method and a moment estimation method, the method comprises the following steps: constructing a gene-environment interaction score test statistical magnitude obeying chi-square distribution, and then applying the gene-environment interaction score test statistical magnitude to judge whether an interaction effect exists in a gene sequence and environment information, so that the method can be used for quickly identifying the influence of multiple pieces of environment information on the gene of the gene-environment interaction site, and can be used for quickly identifying the gene of the gene-environment interaction site. The method is suitable for gene-environment interaction correlation rapid detection of large samples.

The invention discloses a closed-loop system tiny fault detection and estimation method based on data driving, and mainly relates to the technical field of fault diagnosis. The method comprises the following steps: S1, selecting a plurality of working points for a system; S2, calculating P *, T *, b *, [mu]*, [sigma]* and [lambda]*, and selecting a principal component space; S3, determining a fault detection threshold by using an approximate chi-square distribution hypothesis; S4, performing initialization; S5, when a new sampling value is obtained, recording the new sampling value as xk + 1, n, calculating a mean value, and then calculating a score vector value; S6, calculating a mean value and a variance update value of a score vector; S7, calculating the KL distance Kn (tf, t*) of different score vectors of the working point n; S8, estimating the fault amplitude of the working point n; S9, letting k = k+1, and returning to step S5; and S10, judging the fault type of each working point through fuzzy clustering fault diagnosis. The plurality of working points of system operation are fully considered, data in the working range of the system are fully utilized to carry out fault diagnosis on each working point, and the accuracy and robustness of fault diagnosis are improved.

The invention relates to a radar target RCS statistical modeling method based on mixed normal distribution. The method has the advantages that the second-order mixed normal distribution is applied to the RCS statistical modeling; compared with classical RCS fluctuation statistical models (chi square distribution and logarithmic normal distribution), the method can achieve optimal fitting precision for RCS of different target types, different motion feature parameters and different electromagnetic wave parameters, so that the modeling method is universal for describing the distribution features of RCS.