66 results about "Bayesian statistics" patented technology

Exemplary quantitative susceptibility mapping methods, systems and computer-accessible medium can be provided to generate images of tissue magnetism property from complex magnetic resonance imaging data using the Bayesian statistical approach. The likelihood is constructed directly using the complex data. A prior is constructed from matching structures or information content in known morphology. The quantitative susceptibility map can be determined by, e.g., maximizing the posterior. Thus, according to the exemplary embodiment, system, method and computer-accessible medium can be provided for determining information associated with at least one structure. Using such exemplary embodiment, it is possible to receive signals associated with the structure(s), where the signals can include complex data that is in the complex domain.

The related network abnormal detection method based Bayes statistical model comprises: 1. grasping TCP/IP flow data package by bypass interception way; 2. decomposing attributes to form a data matrix; 3. mining data to build normal training matrix, known-abnormal training matrix, and unknown-abnormal training matrix; 4. continual grasping the TCP/IP package to detect them with Bayes evaluator; and 5. alarming the abnormal condition and filling into the known-abnormal training matrix by self-learning way; or else, back to step 4. This invention overcomes defects in prior art.

A pulse oximeter system comprises a data processor configured to perform a method that combines a sigma point Kalman filter (SPKF) or sequential Monte Carlo (SMC) algorithm with Bayesian statistics and a mathematical model comprising a cardiovascular model and a plethysmography model to remove contaminating noise and artifacts from the pulse oximeter sensor output and measure blood oxygen saturation, heart rate, left-ventricular stroke volume, aortic pressure and systemic pressures.

Various embodiments of the invention provide a function centric data system that reduces avionics system weight and power requirements. In some embodiments, the function centric data system is housed in a vibration resistant package. A variety of functions typically performed by other avionics systems are incorporated into the system, allowing centralize power and processing management, reducing weight and improving system reliability. In some embodiments, the function centric data system is configured to provide high rate data sampling, allowing ground stations to apply sophisticated failure prediction algorithms, reducing maintenance costs and mean time between flights. Embodiments include methods of wireless networking with automatic hand offs and adaptive multi-hop topologies to allow this data to be promptly transferred when the aircraft lands. Embodiments also include methods for data processing to predict imminent failures using Bayesian statistics and catastrophe prediction methods.

A thorough understanding of aircraft operations counts at airports is helpful due to the use of those counts in the planning and design process and in the allocation of funds for improvement. Methods of counting aircraft operations at airports lacking full-time personnel are typically based on conventional statistical sampling using relatively small sample sizes due to the inherent difficulty and expense of positioning acoustic or pneumatic counting devices at those airports for extended periods of time. Such methods are often inaccurate because of the lack of sufficient representative samples. A means of counting operations using a combination of Mode C, Mode S, and ADS-B extended squitter aircraft transponder data received using a 1090 MHz software-defined radio system is disclosed herein. The increasing presence of such signals in both controlled and uncontrolled airspace around airports, due to a recent federal mandate that all aircraft in certain types of controlled airspace be equipped with ADS-B Out capability by 2020, lends itself to the measurement of operational parameters associated with the related aircraft. The 1090 MHz signals are received passively; i.e., there is no interrogation from the field-deployed device. The resulting sample counts are typically larger than those determined through conventional data collection procedures. In one aspect, these sample counts are applied to a Bayesian statistical estimation technique, which produces an improved estimate of operations.

The invention provides a novel signal detection scheme aiming at a 60GHz millimeter wave non-linear communication system, wherein the scheme is based on the Bayesian statistical inference mechanism, and capable of effectively solving the problems of system nonlinear distortion and frequency selective multipath fading, and achieving united blind estimate of channel gains and source signals. The method of the detection of the signal of the 60GHz millimeter wave communication system designs an important function applied to the nonlinear system and therefore overcomes the limits of the nonlinear characteristics to the traditional bayes method, further approaches an actual probability distribution function (PDF) through a series of dispersed particles with weights based on the thoughts of the Monte Carlo sequential importance sampling (MC-SIS), and finally utilizes the particle filtering technology to achieve real-time estimating and multi-channel iteration replacement of code element signals (shown in a attached diagram). The method and the device of the detection of the signal of the 60GHz millimeter wave communication system can be applied to the detection of signals of the nonlinear system, improve transmission performance of the system, need no training sequence, and at the same time can achieve the real-time estimating and detection of the signals.

The invention discloses a multi-label learning design method based on a hashing method. Through the combination of a hashing algorithm and a multi-label learning algorithm based on Bayesian statistics, the correlation between labels is effectively utilized so as to improve the predicting performance of a multi-label learning model, labels and neighbors of the labels are introduced to computation of the posterior probability through the characteristics of the neighbors, the correlation between the labels is fully considered, and the accuracy of the algorithms is improved; the problem that the label space in multi-label learning of large-scale data is higher in dimension and sparse is solved through an MinHash algorithm; the purpose of learning large-scale data is achieved by finding the neighbors through locality sensitive hashing (LSH), the neighbors can be rapidly and efficiently found, and the expandability of the multi-label learning algorithm is improved.

A special vehicle identification method for a driverless vehicle comprises the following steps: performing multi-channel information acquisition, performing vehicle detection and characteristic extraction, and performing special vehicle identification. According to depth learning, characteristics of a target vehicle are extracted from multi-channel real-time traffic video and audio information, and by virtue of Bayesian statistical learning, multi-modal integrated special vehicle identification is achieved. According to a logo of the special vehicle and unique image characteristics thereof aswell as acousto-optic characteristics of an alarm lamp and an alarm, the special vehicle and the working state thereof can be quickly identified. By the special vehicle identification method for the driverless vehicle, detection and identification of the driverless vehicle on the special vehicle carrying out a task on a road can be achieved, an operating system of the driverless vehicle is promptly notified of taking avoiding measures, and a decision making basis is provided for the driverless vehicle to timely avoid a road participating object with a traffic priority right; according to the special vehicle identification method for the driverless vehicle, the problem of false identification of the special target vehicle can be reduced, so as to improve the intelligent level of the driverless vehicle.

A method and apparatuses to extend the time interval between out-of-service, in-tank inspections while insuring structural integrity using a risk-based, Bayesian statistical approach comprised of a passing leak detection test and the using the results from (1) tank floor thickness measurements, (2) prior out-of-service tank floor inspection results, and/or (3) acoustic emission corrosion maps of the tank floor to estimate the minimum thickness and maximum corrosion rate of the tank during the extension period. The present invention uses an in-tank, mass-based leak detection system to establish tank integrity, three or more ultrasonic (UT) thickness measurement sensors for measurements of the tank floor at one location, and to establish the spatial distribution of corrosion of the tank floor, one or more prior API 653/12R1 or STI SP001 tank floor thickness inspections and/or three or more in-tank AE sensors mounted inside the tank with vertical and horizontal locations in an oblique plane relative to the tank floor.

The invention discloses an industrial process fault classification method based on self-adaption integrated semi-supervision Fisher discrimination. The method comprises the steps of when off-line modeling is conducted, firstly conducting off-line modeling on unlabeled data, and constituting a semi-supervision random training subset by combining labeled data with the unlabeled data; when iteration training is conducted on a sub classifier each time, conducting semi-supervision Fisher dimensionality reduction to obtain a Fisher discrimination matrix, and obtaining a posterior probability matrix, a combined weight of the sub classifier and a sample weight of the labeled data during next time iteration with the labeled sample data after dimensionality reduction according to a Bayesian statistics method; adopting the posterior probability matrix of the labeled data and a label of the matrix as a training set of a fusion algorithm K near neighbor; during online classification, calling each sub classifier to obtain the posterior probability matrix of an online sample to be detected, and inputting the posterior probability matrix into a fusion K near neighbor classifier with the weight to obtain a final result. Compared with an existing method, the industrial process fault classification method based on the self-adaption integrated semi-supervision Fisher discrimination improves the fault classification result of an industrial process, and more facilitates automated implementation of the industrial process.

The invention discloses an integrated semi-supervised Fisher's discrimination-based industrial process fault classifying method. In the method, offline modeling is first conducted; non-labeled data is randomly sampled and together with labeled data form a plurality of random training subsets; then semi-supervised Fisher dimensionality reduction is conducted to acquire a plurality of Fisher's discrimination matrixes; sampled data with dimensionality reduction is operated according to a Bayesian statistics method to acquire a series of posterior probability matrixes; the posterior probability matrixes of the labeled data and corresponding labels work as training samples adjacent to a measurement layer fusion algorithm K; during online classification, above semi-supervised Fisher's discrimination classifiers are called to acquire a posterior probability matrix of each online to-be-measured sample; and then the posterior probability matrix is input to a measurement layer fusion K adjacent classifier to acquire a final fault classification result. Compared with other methods, industrial process fault classification effect can be improved, knowledge and operation confidence to the process can be enhanced for operators and automatic implantation of the industrial process can be facilitated.

The invention discloses a single nucleotide polymorphism (SNP) site identification method based on digestion library-establishing and sequencing and bayesian statistics. The method is used for processing RAD (restriction site associated deoxyribonucleic acid) sequencing data, searching candidate SNP on an RAD sequencing fragment, and identifying the SNP reliability by employing a bioinformatics analysis method based on bayesian statistics. The method can be used for model and non-model organisms to eliminate the limitation that lots of species are lack of reference sequences and reduce the sequencing cost, and can be used for solving the bottleneck that a reliable statistical method is absent in the process of performing SNP identification by utilizing the RAD data at present, so that the obtained SNP site accuracy is greatly improved.

The invention discloses a Bayesian statistical traceability method for discharging industrial waste water exceeding the standard of sewage pipe network, It includes: 1. Random generation of the initial point (img file = 'DEST_PATH_IMAGE002. TIF' wi= '17' he= '19'/) in the range of the prior information of the unknown parameters; 2, simulate that time series of pollutant concentration of the current parameter (img file = 'DEST_PATH_IMAGE004. TIF' wi= '16' he= '18'/) correspond to the monitoring point, The posterior probability density of unknown parameters (img file = 'DEST_PATH_IMAGE006. TIF'wi= '45' he= '20'/) was obtained by comparing with the actual monitoring data. 3, generate candidate parameters accord to that suggested distribution (img file= 'DEST_PATH_IMAGE008. TIF' wi= '17' he='15'/), (img file= '928204DEST_PATH_IMAGE008. TIF' wi= '17' he= '15'/), The posterior probability density of unknown parameters (img file = 'DEST_PATH_IMAGE010. TIF' wi= '47' he= '19'/) is obtained bycomparing the likelihood degree with the actual monitoring data, 4, extract a random number (img file = 'DEST_PATH_IMAGE012. TIF' wi= '11' he= '13'/), jud whether that candidate value is accepted ornot, outputting an accepted value and a posterior probability density; 5, repeat steps 3 and 4 until that iteration is complete. The invention has the advantages of effectively narrowing the value range of unknown parameters, utilizing the characteristics of the MCMC sampling method, reducing the workload and the sampling time under the condition of ensuring the rationality of the sampling, and improving the traceability efficiency.

The invention discloses a method and a system for dynamically monitoring the software service quality based on Bayesian inference. The method comprises the following steps of: setting null hypothesis and alternative hypothesis, selecting a prior distribution function, reading a training sample, pre-processing the training sample, counting the quantity of samples meeting attributes, and updating a sample set; reshaping the total quantity of samples, the quantity of successful samples and a standard value; and calculating a Bayesian factor, and analyzing, storing and returning a monitoring result. The system comprises a controller, an observer and an analyzer, wherein the controller is used for acquiring the service statement of software, generating an analyzer of different task objectives, transmitting a service standard required to be matched to the analyzer, issuing a command to a data acquisition end, and controlling periodic acquisition; the observer is used for periodically acquiring needed service data from a data service end, and screening, classifying and rearranging information; and the analyzer is used for matching data to form a sample which can be pre-processed, transmitting into a history database for storing, analyzing data by adopting a Bayes statistical module, and storing a result into a monitoring result database.

A pulse oximeter system comprises a data processor configured to perform a method that combines a sigma point Kalman filter (SPKF) or sequential Monte Carlo (SMC) algorithm with Bayesian statistics and a mathematical model comprising a cardiovascular model and a plethysmography model to remove contaminating noise and artifacts from the pulse oximeter sensor output and measure blood oxygen saturation, heart rate, left-ventricular stroke volume, aortic pressure and systemic pressures.

The invention relates to a preceding vehicle detection method based on Bayesian statistical decision. A preceding vehicle detection system comprises a controller disposed on a host vehicle, a single-line laser radar and a camera which are disposed at the front end of the host vehicle and which are electrically connected with the controller. The controller includes a laser radar information processing module, a machine vision information processing module, and a radar visual information fusion decision module. The radar visual fusion decision module includes secondary information fusion which comprises firstly establishing a time-space correspondence model of radar visual information so as to lay the foundation for establishing a visual region of interest; and secondly establishing a vehicle verification function to perform Bayesian decision based on a minimum posterior risk principle to finally verify the existent accuracy of a preceding vehicle. Compared with a traditional preceding vehicle detection system based on laser radar and machine vision, the preceding vehicle detection method of the present invention adds the Bayesian statistical decision based on a minimum posterior risk criterion to the information fusion module so as to effectively reduce a false alarm rate and improve the preceding vehicle detecting accuracy of an intelligent system.

Provided are a device, a method, and a program which allow learning models to be appropriately evaluated or trained. The evaluation device according to an aspect performs the steps of: (A) obtaining, using checking data, a first execution result based on a first learning model as an exemplar model; (B) obtaining, using the checking data, a second execution result based on a second learning model; (C) determining whether or not the first and second execution results satisfy a logical formula; and (D) comparing, using a Bayesian statistical model checking method, respective behaviors of the first and second learning models with each other on the basis of a result of the determination in the step (C).

The invention belongs to the field of language processing and discloses a medical ancient Chinese sentence segmentation method based on Bayesian statistics learning. According to the medical ancient Chinese sentence segmentation method based on Bayesian statistics learning, two tuples and trituples are also added for characteristic attributes or one-tuple, two-tuple and trituple diversified characteristic attributes are combined to obtain multiple groups of experiment data results based on a naive Bayesian method for sentence identification, and finally a best model is obtained; thus, an ancient Chinese sentence segmentation task is achieved. The medical ancient Chinese sentence segmentation method is combined with actual processing text contents, values F of various characteristics in the prior art can be improved by at least 25% by adopting the experiment method, medical ancient Chinese text sentence identification rules are systematically analyzed and concluded, the processing method can be applied to the field of actual traditional Chinese medicine, a medical ancient Chinese text sentence identification corpus is established, and accordingly achievements in scientific research can play a greater role.

The invention discloses a method for predicting transmission line icing thickness, which comprises the following steps: constructing a difference time sequence by utilizing historical data of transmission line icing thickness; establishing an autoregressive metering model for the difference time sequence; estimating a model parameter by a Bayesian statistical inference method; and finally calculating a predicted value of the transmission line icing thickness. In the method, the autoregressive metering model parameter is estimated by the Bayesian statistical inference method so as to predict the transmission line icing thickness dynamically, and the method has the characteristics of high precision and strong real time and solves the technical problem of predicting the transmission line icing thickness under the condition of low-temperature snow and ice storms disasters radically.

The invention discloses a novel spectrum sensing method and device according to the spectrum sensing problem under the condition of an unknown noise variance in practical application. According to the method and device, combined estimation on the state of an authorized user and the noise variance can be achieved. The sensing algorithm is rooted in the Bayes statistical inference principle, the mariginal particle filtering (MPF) technology is adopted, and the unknown noise variance is approached through an iterative estimation mode. The noise variance information obtained through accurate estimation is utilized, the method and device can be conveniently expanded to be applied to (not limited to) a single-node single-antenna sensing system, and the good spectrum sensing performance is obtained (see the attached drawing). The combined estimating and sensing device has good stability, and therefore the device and method can be widely applied to actual cognitive radio system design.

The invention belongs to the technical field of signal characteristic control, and particularly relates to a RCS-based uncertainty model verification method. The method comprises the following steps: 1. an event is defined; 2. an event difference is acquired, and concrete types of an event A and an event B are selected; and sweeping frequency radar cross section data of the event A and the event B under a specific azimuth angle and a pitch angle are extracted, and the difference between the two events is acquired via point-to-point comparison; 3. the expected accuracy D is selected, and a binominal distribution is acquired; 4. the Bayes distribution is calculated so as to acquire uncertainty; 5. confidence is assessed: the interval confidence is accessed in a way that the confidence of Theta dropping into the interval of [pm-d, pm+d] with the difference delta meeting a formula (1.2) is accessed; and 6. the uncertainty calculation formula U(pm, d)=1-Q(pm, d) of the event A and the event B is obtained. The technical scheme can be widely applied to verification of theoretical calculation of various standard models. Bayes statistics has capability of small sample statistics so that a Bayes binomial expression of uncertainty has an advantage in small sample estimation.

The invention provides a financial risk prediction method and device based on Bayesian deep learning and electronic equipment. The method comprises the steps of obtaining a historical user data set, and establishing a training data set and a test data set according to the historical user data set; constructing a Bayesian deep learning model, and performing parameter estimation on to-be-optimized parameters of the deep neural network by using a Bayesian statistical method, the to-be-optimized parameters including weight parameters and bias parameters between layers; training the Bayesian deep learning model after parameter optimization by using the training data set; obtaining user feature data of a target user, and calculating a financial risk evaluation value by using a Bayesian deep learning model; and carrying out financial risk prediction on the target user according to a preset judgment rule and the calculated financial risk evaluation value. According to the method, the generalization prediction capability of the model is enhanced, the prediction precision of the model is improved, and financial risks are further reduced.