116 results about "Probability estimation" patented technology

The invention relates to the technical field of radar remote sensing, discloses a high-order loose body historical activity detection method based on multi-source DEM data, and aims to solve the problem that the historical activity of a high-order loose body in a glacier coverage area is difficult to accurately and quantitatively evaluate in the prior art. The scheme mainly comprises the steps that drainage basin units are automatically divided based on DEM data, and glacier influence areas are screened; multi-source DEM data are fused, and system errors are eliminated through geographic positioning deviation correction, elevation distortion deviation correction and track mode deviation correction; establishing a penetration depth physical model, and correcting the radar penetration depth in the glacier area by using the C / X wave band elevation difference optimization parameter; calculating an elevation change rate by adopting robust regression; and calculating an activity comprehensive index by integrating multiple factors, and identifying a hot spot region and a time period through probability estimation and spatio-temporal clustering. According to the method, automatic and high-precision quantitative evaluation on the historical activity of the high-position loose body is realized, and the method is particularly suitable for geological disaster early warning and risk evaluation in high mountain areas.

The invention discloses a method and a system for cooperatively controlling post-disaster recovery and multi-source resource coordination of three networks, and belongs to the technical field of post-disaster recovery and toughness improvement of a multi-infrastructure system. The method comprises the following steps: obtaining a node importance evaluation function through multi-index fusion, and calculating a recovery priority score of a grey starting power supply; estimating the node state probability to obtain the node state probability; setting boundary conditions of recovery feasibility of the subarea power grid; optimizing a recovery sequence through a grey starting power supply-load-net rack path matching model, and generating a recovery path set; a reinforcement learning algorithm is adopted, node states, resource availability and a path connection matrix are used as state spaces, and an optimal recovery sequence and a resource scheduling scheme are output; and generating a three-network recovery path recommendation, a partition power grid recovery strategy and a multi-element flexible resource scheduling sequence. According to the invention, efficient scheduling and cross-network cooperative control of multi-source flexible resources are realized, and the recovery efficiency and toughness of three networks in a post-disaster scene are effectively improved.

A bot detection service associated with an overlay network operates to score traffic as a probability of being a bot, as opposed to returning a binary classification (i.e., bot or human). According to the approach herein, scoring is determined through probability estimates, wherein a score (the probability) is based on considering a set of detections concurrently. In one embodiment, all (or substantially all) triggered (current) threat detections contribute to the score. The preferred approach penalizes requests that fail all (or substantially all) combinations of detection algorithms. According to a further feature, an automated tuning (autotuning) is also applied, e.g., using real-time empirical statistical models, to adapt the measurement of false positive probability for one or more threat detection algorithms to suit customer traffic trends. The approach herein is also extensible to include any number of future threat detection algorithms.

The invention discloses a defect hidden danger data diagnosis method based on a distribution network SVG single line diagram. The method comprises the steps that basic analysis is conducted on the input SVG single line diagram, basic geometric figure elements are extracted and normalized, a topological connection relation is preprocessed, and a multi-level index structure is constructed; a dynamic topology path reasoning mechanism is constructed, and reasoning and self-adaptive optimization of the topological relation of the electrical equipment are realized through path consistency calculation, topological entropy minimization and a dynamic topology updating strategy; a self-adaptive implicit relational knowledge graph is constructed, and semantic association between the electrical equipment is established through explicit topological relation fusion, implicit function dependence reasoning and dynamic semantic embedding optimization; and constructing a multi-modal time sequence feature vector, calculating the hidden danger risk of the equipment by adopting a space-time defect probability estimation model, and giving an alarm through a dynamic risk early warning strategy to realize prediction and early warning of the hidden danger of the defects of the electrical equipment. According to the method, the problems of semantic understanding, topological reasoning, state inference and defect prediction of the SVG single line diagram are effectively solved.

The invention discloses a sign language recognition model construction method based on uncertainty sample screening and comparative learning, and the method comprises the following steps: S1, obtaining effective motion frames of each sign language vocabulary performed by each sign language performer, constructing an effective motion frame sequence, labeling the effective motion frame sequence to form a data set, and dividing the data set into a training set, a verification set and a test set; s2, inputting the training set into a sign language vocabulary label prediction model for prediction to obtain the sampling probability of each sample; s3, performing sampling probability estimation of a time stage on each sample in the training set to obtain the sampling probability of each sample in the time stage; s4, based on the sampling probability of each sample in the sampling probability time stage based on the sample loss, screening out high-value samples; and S5, constructing a basic recognition model, training the basic recognition model based on calculation results of the steps S2, S3 and S4 by adopting samples in the training set, and obtaining a sign language recognition model.

In one implementation, we propose a bitwise octree coding approach based on deep neural networks and operations on 3D sparse tensors. To encode / decode a certain level of detail (LoD) in an octree, geometric features are first inherited from the previous LoD by upsampling. Then based on the already encoded / decoded voxels, the point cloud geometry is firstly refined by pruning, followed by combining with the known context information. In the end, feature aggregation and probability estimation can be applied to obtain the occupancy probabilities for actual arithmetic encoding / decoding. A corresponding probabilistic training strategy is also proposed for our bitwise octree coding approach.

The invention provides a turbine blade fatigue performance influence factor analysis method and system and a medium, and belongs to the technical field of turbine blade performance prediction.The turbine blade fatigue performance influence factor analysis method includes the steps that a small number of training samples are generated, and a back propagation neural network (BP neural network) describing the relation between the fatigue life of a turbine blade and input variables is constructed; a training sample is added to the sequence, and the BP neural network is updated until the calculated failure probability is converged; based on a failure sample obtained in the process of solving the failure probability, estimating conditional failure probability estimation values at different sample points by using a Bayesian inference theory; and obtaining a fatigue reliability sensitivity estimation value of each input variable according to an average difference between the fatigue failure probability estimation value of the turbine blade and the condition failure probability estimation value of each failure sample point. The method solves the problems that when an existing agent model method is used for solving the reliability and sensitivity of the turbine blade, the sample requirement is large, efficiency is low, and the method cannot be suitable for a high-dimensional problem.

The present invention provides encoding methods and apparatuses, decoding methods and apparatuses, devices, storage media, and computer program products. [Solution] When probability estimation is performed in the encoding method, the probability distribution of the unquantized image features is estimated based on the hyperplier features of the unquantized image features via a first probability distribution estimation network, and then the probability distribution of the quantized image features is obtained through quantization. Alternatively, the probability distribution of the quantized image features is directly estimated based on the hyperplier features of the unquantized image features via a second probability distribution estimation network (a network obtained by performing a simple operation on the network parameters of the first probability distribution estimation network).

The application discloses a wireless sensor network distributed guaranteed probability estimation method under a deception attack, and specifically comprises the following steps: a nonlinear system mathematical model under the condition that innovation is deceived is established; a neural network-based distributed state estimator structure model is established according to the nonlinear system mathematical model under the condition that innovation is deceived; a neural network weight matrix adaptive updating strategy is designed; a distributed guaranteed probability state estimator gain is designed; and estimator parameters are optimized to achieve local optimal estimation performance; the application comprehensively considers the nonlinear characteristics and the influence of malicious attacks existing in actual engineering, can be widely applied to military monitoring systems and environmental monitoring systems, estimates the state of a monitoring or monitoring object, is more in line with engineering application, and is high in practicality.

This application relates to an encoding method, a decoding method, and an electronic device. An example method includes: performing inter prediction on a current frame, to obtain prediction information of the current frame; determining residual information of the current frame based on the prediction information of the current frame and an original picture of the current frame; determining first residual hyperprior information of the current frame based on the residual information of the current frame and prior information of the residual information of the current frame; encoding the first residual hyperprior information of the current frame, to obtain a first bitstream; and performing probability estimation based on second residual hyperprior information of the current frame and prior information of the residual information of the current frame, to obtain a residual probability distribution of the current frame.

The invention relates to a multi-satellite collaborative observation probability estimation method for a window type target. The method comprises the following steps: inputting a statistical time range, an orbital element number and a load observation range of a guiding constellation A, and an orbital element number and a load observation range of a guided constellation B; and determining a to-be-observed area O, based on whether the constellation A is a geostationary orbit satellite, calculating the collaborative observation probability that the constellation A guides the low-orbit constellation B, and outputting a collaborative observation probability result. According to the method, the defect that the time dynamic characteristics of the window type target are not fully considered when the collaborative observation probability of the satellite to the land target is analyzed in the existing method is overcome, and the problem that the window type target collaborative observation success probability is inaccurately estimated in the existing method is solved.

The invention relates to coding equipment and a micro-processing chip. The coding equipment comprises a first data processing module, an arithmetic coding module, a first storage module and a second storage module, the first data processing module is used for reading the to-be-coded image and acquiring to-be-coded pixels in the to-be-coded image and context information corresponding to the to-be-coded pixels; the first storage module is used for storing a context table; the second storage module is used for storing a probability estimation table; and the arithmetic coding module is used for searching the context index value corresponding to the context information in the context table through the first storage module, searching the probability estimation value corresponding to the context index value in the probability estimation table through the second storage module, and coding the pixel to be coded based on the probability estimation value to obtain coded data. According to the invention, hard coding processing of the image data can be realized through the coding device, the coding efficiency can be improved to a great extent, and the problem of low coding efficiency caused by coding through software in related technologies is avoided.

The invention provides a remote sensing image lossless compression method based on overlapped biorthogonal transformation and Transform, and belongs to the technical field of remote sensing image lossless compression coding, and the method comprises the steps: converting a remote sensing image pixel to a frequency domain through overlapped biorthogonal transformation, and obtaining a direct current (DC), a low frequency (LP) and a high frequency coefficient (HP); partitioning the three coefficients, recombining the three coefficients into coefficient tensors, and performing probability estimation on the coefficients by using a probability modeling network to obtain a coefficient estimation probability; and coding according to the estimation probability of the coefficient to obtain a compressed code stream, thereby realizing the lossless compression task of the remote sensing image. According to the method, an overlapped biorthogonal transformation method is utilized, the spatial redundant information of an original image is reduced, three kinds of frequency domain coefficients in concentrated distribution are obtained, accurate probability estimation is carried out on the frequency domain coefficients through strong feature extraction and parameter fitting capabilities of Transformer, efficient lossless compression of the remote sensing image is realized, and the compression efficiency of the remote sensing image is improved. The performance is obviously superior to that of a traditional lossless compression method.

The invention discloses a time-varying thermal failure probability estimation method and system for a deep groove ball bearing, and belongs to the field of bearing design. The method comprises the following steps: constructing a basic parameter model containing a mixed random parameter vector and a limit state function; establishing a transient thermal network model to calculate time-varying temperature response; constructing an agent model to replace high-time-consumption heat generation rate calculation; a self-adaptive time node sampling strategy is adopted, a proxy model and a thermal network model are combined to process large-scale samples, and the time-varying thermal failure probability is estimated. The system comprises a basic parameter modeling module, a transient thermal network simulation module, an agent model construction module and a self-adaptive probability evaluation module which correspond to the system. According to the method, the time-varying and random dual uncertainty is fused, and the proxy model and the adaptive sampling strategy are utilized, so that the defects of conservative property of a traditional static evaluation method and low efficiency of a traditional probability simulation method are effectively overcome, and accurate evaluation of the thermal behavior of the bearing under the dynamic working condition and remarkable improvement of probability calculation efficiency are realized.

The application provides an aircraft intention recognition probability estimation method and system, and relates to the technical field of aircraft intention recognition. Basic information of a target aircraft is acquired, attribute information and state information are determined in the basic information; semantic information fuzzy sets are generated according to the attribute information and the state information to serve as antecedent structure parameters; target task information is acquired, and membership functions in consequent structures are determined according to the target task information; the intention probability prediction model of the target aircraft is generated by combining the basic information and the target task information and through aggregation of preset semantic rules; the basic information of the current target aircraft is acquired, and the flight intention is determined in combination with the intention probability prediction model. Through multiple rules, complex nonlinear relationships are approximated, various complex data are fitted, each rule has semantic characteristics, and strong interpretability can be achieved, so that the technical effect of efficiently and accurately identifying the aircraft intention is realized.

This invention discloses a driver state estimation method based on multi-task learning, comprising: using a shared pre-trained CNN model as the backbone network to extract features from an input target image to obtain shared features for the target tasks; wherein the target tasks include a keypoint prediction task, an occlusion probability estimation task, and a head pose estimation task; inputting the shared features into each task module to complete the corresponding task prediction, obtaining keypoint prediction results, occlusion probability estimation results, and head pose estimation results; and generating a driver state estimation result based on the keypoint prediction results, occlusion probability estimation results, and head pose estimation results. This invention efficiently solves the three problems of driver facial keypoint detection, occlusion probability estimation, and head pose estimation, improving the accuracy of the estimation results and can be widely applied in the field of deep learning technology.

In one implementation, we propose a bitwise octree coding approach based on deep neural networks and operations on 3D sparse tensors. To encode / decode a level of detail (LoD) in the octree, geometric features are first inherited from the previous LoD by upsampling. Then, based on the voxels already encoded / decoded, the geometry of the point cloud is refined first by pruning, followed by combining with known context information. Finally, feature aggregation and probability estimation can be applied to obtain occupancy probabilities for actual arithmetic encoding / decoding. We also propose a corresponding probabilistic training strategy for our bitwise octree coding approach.

The invention discloses an AK-MCS reliability analysis method based on a dynamic critical reinforcement learning function, and belongs to the field of mechanical equipment structure reliability analysis. According to the method, a DC learning function is provided, and the coupling influence of a sample point prediction value and a prediction variance on the failure probability estimation precision is comprehensively considered through an adaptive weight mechanism. Preferentially selecting sample points with large prediction variance in the exploration stage to quickly improve the global approximation capability of the model; in the refinement stage, a learning function is converted to be dominant by a mean item, the selection weight of sample points near a limit state curved surface is dynamically increased, local fine description of a critical region is enhanced, and a global relative uncertainty index is introduced to realize smooth transition from global exploration to local refinement. According to the method, the problems of excessive iteration and local optimization of a traditional learning function are effectively avoided, the construction efficiency of the Kriging proxy model and the robustness of failure probability estimation are remarkably improved, and the method is suitable for efficient reliability analysis of a complex mechanical equipment structure.

A method to provide a model-assisted inline wafer-level inspection during high volume manufacturing is disclosed. More particularly, a method for using a computational model to generate fingerprint wafer defect maps and then guide wafer selection for inline inspection is disclosed. A computational probability prediction model is disclosed to generate defective die probability estimates with improved accuracy and versatility to guide different wafers for inspection.

The present application relates to the technical field of image classification, and particularly relates to a training method and device of an image classification model, equipment and a medium. The method obtains an auto-encoding feature vector and a reconstructed image feature vector through an unsupervised image reconstruction model, obtains K feature vector clustering clusters of the auto-encoding feature vector, determines a noise probability transition matrix for representing noise confusion information between various image categories, obtains a category probability estimation vector according to a supervised image classification model, determines model loss by combining the noise probability transition matrix to perform model training, obtains a trained supervised image classification model to perform image classification, measures the distribution of noise labels and non-noise labels through the feature vector clustering clusters, obtains the noise probability transition matrix to represent the noise confusion information between the image categories, and measures the loss of the supervised image classification model, reduces the influence of noise labels on the supervised image classification model, and improves the accuracy of the supervised image classification model.

Systems and methods are configured for accessing data representing video content, the data comprising a set of one or more symbols each associated with a syntax element; performing a probability estimation, for encoding the data, comprising: for each symbol, obtaining, based on the syntax element for that symbol, an adaptivity rate parameter value, the adaptivity rate parameter value being a function of a number of symbols in the set of one or more symbols; updating the adaptivity rate parameter value as a function of an adjustment parameter value; and generating, based on the updated adaptivity rate parameter value, a probability value; generating a probability estimation; and encoding, based on the CDF of the probability estimation, the data comprising the set of one or more symbols for transmission.

A method of training a computational defect probability model to guide an inspection using an inspection tool is disclosed. More specifically, a method of combining input data from a plurality of devices, correlating metrology data with defect data from devices processed according to the same device manufacturing technology to train a computational defect probability, and applying the trained computational defect probability model to a guided inspection is disclosed. A computational defect probability prediction model is disclosed for generating defective die probability estimates in a shorter time with improved accuracy and versatility to guide inspection.

The invention discloses an Internet of Things terminal network security early warning platform based on artificial intelligence, which performs cross-domain feature alignment on information domain and physical domain data analyzed at a bottom layer, and innovatively converts the information domain and physical domain data into a time sequence texture grey-scale map, thereby utilizing the characteristics of a convolutional neural network on a spatial receptive field, and improving the security of the Internet of Things terminal network security early warning platform. And deeply extracting potential semantic association and time sequence evolution laws among the heterogeneous data streams. Finally, through full-connection classification and real-time estimation of threat probability, a full-link early warning closed loop from multi-dimensional feature fusion, deep feature mining to defense instruction automatic generation is realized, and the security defense efficiency and response speed of the industrial-grade Internet of Things in the face of advanced persistent threats are greatly improved.

To provide an estimation and control method for frequency errors of extreme events.SOLUTION: The present invention proposes a frequency shift functional and a frequency shift functional estimation formula based on probability estimation, and constructs a method for reversely solving a confidence frequency shift value epsilon according to the frequency shift functional to improve a frequency distribution function. The present invention is suitable for estimating and controlling errors of occurrence frequency prediction of extreme events in various frequency statistical models by using hydrological observation data under probability statistical significance, so that the prediction accuracy and reliability of hydrological extreme events such as storm surge, rainstorm flood, tsunami, and the like are improved, risks caused by extreme natural disasters are reduced, and loss of life and property that may be caused is reduced.SELECTED DRAWING: Figure 1

This work classification estimation method includes: a step for acquiring the likelihood of a candidate classification from a likelihood model that is a machine learning model; a step for acquiring the appearance probability of the candidate classification from a probability model that is a statistical model; a step for estimating the classification of work executed by a caregiver, on the basis of the likelihood and the appearance probability of the candidate classification; and a step for outputting the results of estimating the classification of the work.

This invention relates to the field of image classification technology, and more particularly to a training method, apparatus, computer device, and medium for an image classification model. The method acquires images to be classified and their category labels from a training set; obtains a category probability estimation vector for the images to be classified based on an image classification model; determines a category probability prediction vector; calculates a first model loss based on the category probability estimation vector and category labels; determines a loss adjustment parameter based on the number of iterations and a first mapping relationship between the number of iterations and the loss adjustment parameter; calculates a second model loss based on the category probability estimation vector, the category probability prediction vector, and the loss adjustment parameter; calculates the total model loss based on the first and second model losses; repeatedly iterates and trains the image classification model; and dynamically adjusts the second model loss using the loss adjustment parameter, thereby improving the reliability and accuracy of the second model loss and ultimately improving the accuracy of the image classification model.

When probability estimation is performed in encoding and decoding processes, probability distribution of an unquantized image feature is estimated based on a hyperprior feature of the unquantized image feature via a first probability distribution estimation network, and then probability distribution of a quantized image feature is obtained through quantization. Alternatively, probability distribution of a quantized image feature is directly estimated based on a hyperprior feature of an unquantized image feature via a second probability distribution estimation network.

The invention discloses a ground cluster target damage probability estimation method based on a combined damage matrix. The method comprises the steps of 1, calculating damage matrix parameters; considering that a cluster target is composed of seed targets and ammunition co-species for hitting the cluster target, calculating a damage matrix of the seed ammunition to the seed targets, and forming a set of complete damage matrix parameters; 2, a cluster target coordinate system is set, and the number, types, positions and postures of sub-targets are given; step 3, setting ammunition quantity, types and end point parameters; and 4, calculating the damage probability of the selected D ammunitions to the cluster target by adopting a Monte Carlo sampling method. The method can be suitable for rapid calculation of the damage probability of a cluster target formed by any space combination of multiple sub-targets, a damage matrix combined application framework is established, and the application scene of the damage matrix method is expanded to the cluster target from a single target.

To provide an encoding method and a decoding method for arithmetically encoding a sequence of information values into an arithmetic coded bitstream using providing the bitstream with entry point information making it possible to resume arithmetic decoding of the bitstream after a predetermined entry point.SOLUTION: An encoding method comprises: symbolizing 101 information values into symbol strings so as to acquire a sequence of symbols; and arithmetically encoding 102 the sequence of symbols. The arithmetic encoding comprises: subdividing 103 a current interval, which defines a current version of a coding state of an arithmetic encoder, for each symbol, according to a probability estimate for the respective symbols; selecting 104 a subinterval out of a plurality of subintervals according to a symbol value of the respective symbols; and renormalizing 105 encoder-internal parameters which define the coding state under continuing a bitstream.SELECTED DRAWING: Figure 1

To provide an encoding method for arithmetically encoding a sequence of information values into an arithmetically coded bitstream, using providing entry point information to the bitstream, thereby allowing arithmetic decoding of the bitstream to be resumed forward from a predetermined entry point, and a corresponding decoding method.SOLUTION: An encoding method includes the steps of: symbolizing information values into symbol strings so as to acquire a sequence of symbols (step 101); and arithmetically encoding the sequence of symbols (step 102). The step of arithmetically encoding includes the steps of: subdividing a current interval that defines a current version of an encoding state of an arithmetic encoder according to a probability estimate for the respective symbol for each symbol (step 103); selecting a subinterval out of a plurality of subintervals according to a symbol value of the respective symbol (step 104); and renormalizing encoder-internal parameters which define the encoding state under continuing the bitstream (step 105).SELECTED DRAWING: Figure 1