174results about How to "Reduce relevance" patented technology

The invention discloses a driving intention identification method based on an improved HMM and SVM double-layer algorithm. The method includes: step 1, classification of driving intentions of drivers: dividing the driving intentions of the drivers into five categories: emergent left lane-changing, normal left lane-changing, lane maintaining, normal right lane-changing, and emergent right lane-changing; step 2: experiment data acquisition and processing; step 3: offline training of the improved HMM and SVM double-layer algorithm; and step 4: identification of the driving intentions of the drivers. According to the method, a human-vehicle-road system is completely considered, information of vehicles, roads and drivers is acquired, the HMM and SVM double-layer learning model is employed, and the accuracy and the timeliness of a driver lane-changing intention identification model are improved.

The invention relates to a heart valve prosthesis. The heart valve prosthesis comprises a stent and artificial valves; the stent comprises an atrium section and a valve suturing section, the artificial valves are fixedly connected to the valve suturing section, a skeleton of the far end part of the valve suturing section extends to form a supporting section, and the supporting section and the valve suturing section are of an integrated structure; one end of the atrium section is connected with the far end of the supporting section, the other end of the atrium section is free and is located on the cross section perpendicular to the central axis of the artificial valves, the valve suturing section is located in the supporting section, the center of the valve suturing section does not coincide with that of the supporting section, the cross section area of the supporting section is larger than that of the valve suturing section, the cross section area of the atrium section is larger than that of the supporting section, the cross section area of the valve suturing section is smaller than that of valve rings of autologous valves of a patient, and in a free state, at least part of the supporting section is located among the valve rings of the autologous valves of the patient.

The invention relates to an organic light-emitting display device capable of preventing adjacent pixels from being shorted out. The organic light-emitting display device comprises a substrate and a pixel defining layer, wherein the pixel defining layer forms a continuous grid structure; pixel units are formed in the continuous grid structure; and an isolating structure for preventing the adjacent pixel units from being shorted out is arranged in the pixel defining layer. According to the organic light-emitting display device provided by the invention, by the isolating structure arranged on the pixel defining layer, the problem that an existing OLED device easily generates a transverse short circuit is effectively solved; and the existing device efficiency can be kept.

The invention relates to a compressive sensing image reconstruction method based on a pseudo-inverse adaptive matching pursuit (PIAMP), which is characterized by comprising the steps that a gauss random matrix is selected for projecting an image to form a measurement signal matrix; an image sparse base is constructed; each column of the measurement signal matrix is reconstructed by adopting the PIAMP to form a sparse coefficient of each column of the image; and the sparse coefficients of the columns of the image are arrayed and subjected to cosine transformation to form a reconstruction image. The method particularly takes account of the dictionary relevance, iteration stage division and the like in a sparse coefficient reconstruction process, so that selection of the sparse coefficients is more accurate, and the precision of image reconstruction is higher.

The invention discloses a shadow detection method based on an attention mechanism. The shadow detection method comprises the steps of obtaining a to-be-processed public shadow data set for a trainingsystem; obtaining a to-be-detected shadow image from camera equipment or a local hard disk; preprocessing the public shadow data set and obtaining a corresponding training set and a corresponding testset; preprocessing the to-be-detected shadow image; constructing and fusing all modules of a system core neural network to form a convolutional neural network based on an attention mechanism; calculating network prediction and label loss, and adjusting network parameters according to the loss; completely training the deep convolutional neural network and inputting a preprocessed to-be-detected shadow image into the deep convolutional neural network; and outputting a shadow detection result, and performing shadow attribute classification on each pixel to complete a shadow detection process ofthe custom data. According to the invention, the shadow feature extraction capability is improved, the semantic relevance is reduced, the generalization and high efficiency of the detection system areimproved, and the shadow detection result is more accurate.

The invention discloses an antenna multiplexing device. The antenna multiplexing device comprises a Cellular main antenna, at least one Cellular diversity antenna, a WLAN antenna, a first radio-frequency switch, a second radio-frequency switch, a first communication module, a second communication module and a gating module; one end of the first radio-frequency switch is connected with the first communication module; one end of the second radio-frequency switch is respectively connected with the first communication module and the second communication module; the second communication module comprises a multiplexing port and an antenna connection port; and the second communication module is connected with the second radio-frequency switch through the multiplexing port and connected with the WLAN antenna through the antenna connection port. According to the antenna multiplexing device and the mobile terminal provided by the invention, the main antenna and the WLAN antenna form a MIMO antenna in a first working mode through the gating module; therefore, the relevancy of the MIMO antenna in multiple frequency bands can be reduced; and the antenna efficiency can be increased.

The invention discloses a hypersonic aircraft self-adaptive learning control method based on a deviator characteristic model. Two variable corrected parameters are introduced, and by properly regulating the corrected parameter, the control increment at each step is limited so as to solve the problem that an executive mechanism rudder shakes too quick. The method comprises the following steps of: establishing the deviator characteristic model for laying a foundation for a golden section self-adaptive learning control law; providing a dual-parameter distinguishing method for reducing a transient error of the system control; and providing a golden section self-adaptive learning control design method based on the deviator characteristic model, thereby limiting the size of the control increment by regulating the control law corrected parameter, and reducing the relevance of a characteristic model coefficient and a state, and solving a coefficient distinguishing problem of the characteristic model. Aiming at the hypersonic aircraft, the golden section self-adaptive learning control law and the dual-parameter distinguishing algorithm based on the deviator characteristic model are designed to reduce the transient error and the shake of the rudder.

The invention provides an Android malware detection method based on an improved Bayesian algorithm. The feature attributes of Android malicious programs and well-behaved programs are analyzed and classified through the improved Bayesian algorithm to realize the malware detection method based on the improved Bayesian algorithm. A judgment on whether software is malware is implemented from the aspect of permission application of applications. According to the method, a permission request label in an Android permission request mechanism is taken as a detection data source. The malware and well-behaved software are distinguished in a permission request label combination way, and a detection model is built by using the improved Bayesian algorithm. The improved Bayesian algorithm is characterized in that mutual independence among attributes of the data source is considered, and a naive Bayesian classifier is used for performing data modeling, so that the detection index is increased greatly, the detection accuracy is increased, and the false alarm rate is lowered.

The present invention relates to an inter-network data header compression realization method. According to the method, two embedded ARM (Acorn RISC Machine) processors are adopted as a hardware basis, and the kinds of hardware are designed into gateways in different network domains by means of gateway data forwarding; a multi-processing mechanism is adopted to perform classification processing on the different data packet types of the different gateways; and a network address and physical address mapping method is adopted to achieve host communication between different local area networks. The inter-network data header compression method can be applied to an OSI (Open System Interconnection) seven-layer model-based network protocol. Different from a traditional a third-layer switching technology, the inter-network data header compression method of the present invention uses the network address and physical address mapping method in the ARP (Address Resolution Protocol) technology to process the data packets of a data link layer, so that the gateways can directly forward the data after receiving data without making the data processed by the upper layer of a network protocol, and therefore, the forwarding efficiency of the data can be increased by 15%; the quantity of data transmission in a communication link can be reduce by 12%; and good anti-jamming performance can be realized when video voice data packets are transmitted.

A continuous regenerative flue gas residual heat recycle device belongs to the industrial furnace energy saving field. The device is made from high heat resistance and conduction materials. A gas passage in the device is separated into mutually independent multistory spaces. Flue gas passages and air (gas) passages are arranged in a stratified and staggered way (that is a layer of flue gas passage is connected with a layer of air (gas) passage in the height direction, a staggered arranged sandwich structure). Every layer of gas flow passage is a honeycomb channel, which is made from high thermal conductivity materials. The hest recycle device has the advantages mentioned below. First, the heat exchange rate is fast. The thermal conductivity rate of a regenerator is high. And heat exchange between flue gas and air can be done rapidly. Second, the device can work continuously. Communication is not required in the wind supply and the smoke exhaust. And the system is simple. Third, utmost waste recovery is really realized. No heat and fuel loss caused by volume effect is made during the communication. Fourth, with a long service life, the device can be flexibly arranged. And the device has relative small relevance with the furnace, which makes the failure risk lowered and is convenient for the engineering management.

The present invention relates to the communication field, and discloses a data framing method and equipment thereof, which reduces the probability of synchronization error. In the invention, M verifying blocks are obtained by coding N data blocks, wherein each data block contains a data synchronization header, each verifying block contains a verifying synchronization header, the data synchronization header and the verifying synchronization header are different, the N data blocks are divided into X parts, and the M verifying blocks are divided into Y parts (N>=X>=1, M>=Y>=1, at least on of X and Y is greater than 1, the difference of the absolute values of X and Y is less than 2). Then, various parts of the data blocks and verifying blocks are arranged with an interval between each other to form a frame.

The invention provides an LDAP-based distributed cluster management and monitoring system which can be used by the cluster manager to control and monitor the operational status of every server. The system presents as an inverted tree, wherein the root node represents the management end, and the management end presents as a sub-tree and can also include a plurality of sub-trees to represent system data. Firstly the information of every server is self-organized and preserved in a local tree and every node of the tree is an information body representing a class of information which is expressed by a series of 'attribute-value'; the information can be preserved in any mode like database. LDAP simplifies the access by adding an abstraction layer to the data from an angle which is easy for people to understand.

The invention provides an operation switch device internal defect determination method based on SF6 gas decomposition products, comprising steps of using a gas chromatography detection system to detect the SF6 gas decomposition products of the operation switch device, performing statistic on the detection results of the SF6 gas decomposition products to determine the characteristic gas of the decomposition products and the characteristic parameter of the characteristic gas, and determining the internal defect type of the operation switch device according to the characteristic vector. Compared with the prior art, the invention establishes the characteristic parameter determination method for the SF6 gas decomposition products under various device faults and improves the reliability and the effectiveness of the switch device internal defect determination.

The invention discloses an intellectualized light source linear array CCD liquid level measurement method and a measurement device. By using a feedback measurement structure, the method can obtain a result of comparably high precision through measurement of twice or a plurality of times and the intellectualized regulation of the light source. The device mainly consists of a point-to-point switch controllable linear array light source, a linear array CCD, a control circuit, etc. During the measurement of liquid level, all the luminous points on the linear array light source intelligently complete primary measurement scanning and determine the basic position of liquid level; the partial luminous points on the linear array light source are closed so as to complete the accurate measurement of liquid level. The intellectualized light source linear array CCD liquid level measurement method is a liquid level measurement method with higher precision and also an optical liquid level measurement method needing no optical imaging lenses, and has the advantages of little effect on the precision by the temperature change and concentration change of liquid, relatively simple structure, easiness for arrangement and the capability of working under corrosiveness and other vile industrial environments, etc. If a receiving device is arranged outside a transparent container, non-contact measurement can be realized.

The invention discloses a bidirectional TVS and a manufacturing method of the bidirectional TVS. The bidirectional TVS comprises a first conduction type substrate, a second conduction type base region, a first conduction type conducting layer, a plurality of grooves, oxide layers, a first electrode and a second electrode, wherein the second conduction type base region is formed on the first conduction type substrate, the first conduction type conducting layer is formed on the second conduction type base region, the grooves all penetrate through the first conduction type conducting layer and the second conduction type base region, extend into the first conduction type substrate, and are filled with the oxide layers, the first electrode is formed on the first conduction type conducting layer, and the second electrode is formed on the back face of the first conduction type substrate. According to the bidirectional TVS and the manufacturing method of the bidirectional TVS, lateral side isolation of a PN junction is achieved through a deep groove structure, the grooves are filled with isolation materials, and electric leakage from the lateral side of the PN junction is reduced; the relevance between the widths of the isolation grooves and the depths of the isolation grooves is greatly reduced, and the size of a chip of the bidirectional TVS is easily reduced.

The invention discloses a dynamic searchable public key encryption method, belonging to the technical fields of cryptography and cloud storage. According to the method, functions of generation, retrieval and dynamic updating of ciphertext data in a cloud storage system under a public key encryption mechanism are realized. A sender encrypts key words of a file to generate a searchable ciphertext for (ID, W), a to-be-uploaded file is encrypted to generate a cipertext file with a file ID, and lastly, the searchable ciphertext and the cipertext file are uploaded to a server together. A receiver submits a trap door of a to-be-searched key word W to the cloud storage server, the cloud storage server carries out retrieval to obtain a corresponding searchable ciphertext, the searchable ciphertext is further decrypted to obtain a relevant file ID, and the corresponding cipertext file is found according to the file ID and is returned to a user. The receiver submits a trap door of a to-be-deleted file ID to the server, the server retrieves a cipertext index according to trap door information in order to obtain a corresponding searchable ciphertext index, and then the searchable ciphertext index and the cipertext file corresponding to the file ID are deleted.

The invention relates to the technical field of unmanned aerial vehicles, in particular to a kinetic modeling method and device for a rotor unmanned aerial vehicle. The kinetic modeling method for the rotor unmanned aerial vehicle comprises the steps that flight data of the rotor unmanned aerial vehicle is obtained; parameter estimation is conducted according to the flight data of the rotor unmanned aerial vehicle, and a corrected parameter estimation value of the rotor unmanned aerial vehicle in a six-degree-of-freedom model built through the Newton's law is obtained; a kinetic model of the rotor unmanned aerial vehicle is obtained from the six-degree-of-freedom model according to the corrected parameter estimation value and the flight data. Accordingly, the rotor unmanned aerial vehicle is simulated through the Newton's law, by means of adjustment of kinetic feature expression parameters of any degree of freedom on six degrees of freedom, correction of related parameters can be conducted, and simulation of the kinetic features of the unmanned aerial vehicle is more precise on the basis of simplifying the kinetic features.

A data store has a data array for storing data values and a tag array for storing tag values for tracking which data values are stored in the data array. The associativity of the data array is greater than the associativity of the tag array. This means that fewer tag entries need to be accessed on each data access than in a conventional data store, reducing power consumption.

The invention relates to a communication method and device, and in particular relates to a VxWorks-based lithography machine double-workbench communication method and device, aiming at solving the problem that a conventional VxWorks operating system using a TCP (Transmission Control Protocol) or an IP (Internet Protocol) is poor in communication real-time performance. The communication method and the device are realized based on a double-workbench lithography machine, an upper computer and a VME (Virtual Machine Environment) industrial personal computer, and adopts double channels of the TCP and a UDP (User Datagram Protocol) for communication, wherein the channel of the TCP is used for transmitting instructions of the upper machine, the channel of the UDP is used for transmitting execution result information of the VME industrial personal computer, a buffering queue is used for caching all the instructions and data information which are then read out one by one, and finally executed and sent, and a zero copy buffering area is adopted for nesting word interfaces. The communication method and device are used for communication when a lithography machine double-workbench is arranged in an environment which is provided with the VxWorks operating system and has access to the Ethernet.

The invention discloses an apparatus for overlay error detection and a measurement and calibration method. The apparatus comprises a light source, an illuminating system, an objective lens, a reference mark, a detector, and a rotatable workpiece table, wherein the light source is used for generating incident light; the incident light is scattered light; the illuminating system is used for dividing the incident light into two beams, wherein one beam is used as measurement light to come into the objective lens while the other beam is used as monitoring light to come into the detector; the objective lens is used for enabling the measured light to come into an overlay mark in an inclined manner; the reference mark is used for generating reference light; the detector is positioned on the pupil plane of the objective lens and used for detecting diffraction spectrum of the overlay mark; and the rotatable workpiece table is used for bearing and rotating the reference mark and / or the overlay mark. The reference light, which passes through the light path that is totally the same as that of the overlay signal, is adopted for performing measurement signal normalization, so as to eliminate influence on the overlay signal from asymmetry of transmittance, space distribution non-uniformity of the incident light, space distribution changes, sensitivity non-uniformity of the detector, and the like; and a similar measurement and calibration scheme is never mentioned in the prior art.

The invention discloses a registration and login system based on blockchain zero knowledge proof and an application of the registration and login system in a registration method given by minimizationof information amount, a token number issuing method of a node and a login method with a hidden login state. The registration and login system comprises a token issuing module, an account address generation module, an account registration module and an account login application module. According to the invention, the registration and login system is integrated into the blockchain, and based on thecharacteristics of anonymity, decentration, tamper resistance and the like of the blockchain, and zero knowledge range is added to prove that personal information such as a user account is effectively protected, so that the privacy security of the user is improved, and the privacy information of the user is protected.

The invention relates to a high-resolution remote sensing image feature selection method based on a Cramer's V index. The method comprises the following steps of: 1) carrying out pretreatment and image feature extraction on an obtained remote sensing image; 2) carrying out continuous feature discretization parallel processing based on the Cramer's V correlation index; 3) carrying out parallel processing to obtain two contingency tables between two factures; and 4) carrying out feature selection based on the Cramer's V correlation index. The method is good in feature selection effect, high in efficiency and high in applicability, and is capable of efficiently improving the classification accuracy of the remote sensing image. Besides the remote sensing processing, the method can be widely applied to various problems such as pattern classifications of various high-dimensionality and complex type data sets (such as texts, images, medical diagnosis, and genetic data), data mining and visualizing, and the like.

The invention discloses a modularized design method for a braking system for a vehicle. The modularized design method solves the technical problems that in the prior art, the efficient design of the braking system of the vehicle cannot be met, and in addition, the quality of the designed braking system of the vehicle is not high. The modularized design method comprises the following steps that the braking system used by the vehicle is subjected to modularized classification, and an assembly warehouse of a gas storage tube assembly module is built; an assembly warehouse of a valve element assembly module is built; an assembly warehouse of a pipeline assembly module is built; element assemblies are selected from the built gas storage tube assembly module, the built valve element assembly module and the built pipeline assembly module for arrangement according to a framework model of the whole vehicle, and a three-dimensional model of a currently required braking system is generated. The goals of reducing repeated work to the maximum degree, improving the assembly efficiency during the design and saving the time are achieved.

The invention discloses a crowd sensing dual privacy protection method based on a block chain. The method comprises the steps that a task publishing center publishes a sensing task and sends a messagerelated to the task to a block chain network; the block chain network generates a perception task notification through the smart contract and broadcasts the perception task notification to all workers; the workers execute the sensing task and upload the sensing data to the block chain network; the nodes in the block chain network compete for the accounting right by using a consensus algorithm, the miner obtaining the accounting right verifies the quality of the perception data, quantifies the rewards according to the contribution degree, and then generates a final block; and when the final block is recorded on the block chain, rewards are also allocated to corresponding workers and miners, and the like. The method has the advantages that consensus is achieved through the contribution degree consensus algorithm, the shared block chain is used for storing records, on one hand, transparent records are provided, convenience is brought to auditing, and meanwhile malicious attack behaviorsof lawbreakers on the shared network can be prevented.

The invention relates to asymmetrical heart valve prosthesis. The asymmetrical heart valve prosthesis comprises a stent and an artificial valve; the stent comprises an atrium section, a supporting section and a valve sewing section; the supporting section is connected with a far end part of the valve sewing section; the atrium section is connected with the supporting section, or the atrium section is connected with the far end part of the valve sewing section; the artificial valve is fixedly connected with the valve sewing section; in a cross section which is perpendicular to the central axis of the artificial valve, the centre of the valve sewing section is not overlapped with the centre of the supporting section; the cross-sectional area of the valve sewing section is smaller than that of an autologous valve ring of a patient; in a free state, the valve sewing section is at least partially positioned between autologous valve leaflets of the patient; the cross-sectional area of the atrium section is greater than or equal to that of the autologous valve ring of the patient.

The invention relates to a charge redistribution successive approximation-type analog-to-digital converter and a control method thereof. The charge redistribution successive approximation-type analog-to-digital converter comprises a comparator, a buffer, a control circuit, multiple first capacitors and at least one second capacitor, wherein the comparator is used for generating a comparison result; the buffer is used for storing a digital output code and deciding the bit value of the digital output code according to the comparison result; the control circuit is used for generating a control signal according to the digital output code; the first capacitor is provided with a first endpoint and a second endpoint, and the first endpoint is coupled with the first input end of the comparator; and the second capacitor is provided with a third endpoint and a fourth endpoint, and the third endpoint is coupled with the first input end of the comparator. Before the voltage of the second endpoint of each first capacitor and the voltage of the fourth endpoint of the second capacitor are switched, the second endpoint is coupled with the first voltage and the fourth endpoint is coupled with the second voltage, and the first voltage is not equal to the second voltage.

The disclosed device includes following modules: being connected to module for loading and reading authentication data, the module for parsing nonreal / real time data is in use for parsing nonreal / real time data in authentication data; the data check module is connected to module for parsing nonreal time data, module for parsing real time data and module for supplying auxiliary data; module for calculating parameters is connected to the data check module and the data input module; data output module is connected to the data check module and the module for calculating parameters; the data storage module is connected to data output module. Through interaction among authentication module at software terminal, service-call dialogue control function, and home server of user / authentication center, the method accomplish procedure of authentication. The invention makes software terminal, which does not support application of USIM and ISIM, possible to connect to IMS.

The invention relates to a wireless communication technology, in particular to an anti-interference wireless communication method based on deep reinforcement learning. The method comprises the following steps: using two convolutional neural networks: one convolutional neural network calculates a value function, and the other convolutional neural network performs action selection based on a calculation result of the value function; adopting priority experience sampling in an experience playback stage, so that experience samples with higher priorities are sampled preferentially, updating parameters of the convolutional neural network based on the experience samples, and updating the priorities of all the experience samples through calculation of the updated convolutional neural network; adopting a forward action reservation strategy, designing a Gaussian-like function to judge the value of the current action, and dynamically adjusting and controlling the probability that the current action is continuously executed. According to the method, the optimal sending power and the optimal communication frequency band can be intelligently selected, the learning speed of the whole system is improved, and the optimal sending mode can be learned under the condition that a third-party attacker model is unknown.

The invention relates to a control terminal and control method thereof, a mobile platform and control method thereof, after receiving multi-signal information (S101), the control terminal can construct the multi-signal information into a plurality of sub-multi-signal information units (S102) and transmit the plurality of sub-multi-signal information units to the mobile platform (S103), after receiving the plurality of sub-multi-signal information units (S901), the mobile platform reconstructs multi-signal information according to the received plurality of sub-multi-signal information units (S902), then during a process of transmitting the plurality of sub-multi-signal information units from the control terminal to the mobile platform, if one or more sub-multi-signal information units in the plurality of sub-multi-signal information units have an error code or be lost, the mobile platform still can reconstruct the multi-signal information according to other sub-multi-signal informationunits, and differential positioning can be completed according to the reconstructed multi-signal information, and fault tolerance and reliability of multi-signal information transmission are improved.