58results about How to "Guaranteed time consistency" patented technology

The invention relates to an automatic generation method of an interactive three dimensional city panoramic map. Based on city morphology and gestalt theory, the method combines roads widening, viewpoint raising, building displacement and building height reducing with optimization method through establishing energy function deviating argument and shielding argument for estimating city space cognition and uses linearization constrained condition energy function to solve minimizing energy function in real time, therefore automatic generation of the interactive three dimensional city panoramic map is achieved. The making method of the three-dimensional digital city panoramic map is characterized by fast visualization speed and little deformation of city scenes, having broad application prospect in city navigation, space position service and the like.

The invention discloses a combined lightning monitoring and forecasting system. A lightning alarming instrument is arranged at the right upper end of an L bracket and toward the sky; an atmospheric electric field instrument is arranged at the tail end of the L bracket and toward the ground; transmission terminals of the two instruments are connected with a signal acquisition and processing device which has the following structure that: the atmospheric electric field instrument comprises two paths of output terminals, one path is that an electric field signal of the electric field instrument is sequentially connected in series with a band-pass filter, a phase regulating circuit, an electric field signal amplifier and an isolating circuit, while the other path is that a synchronous signal of the electric field instrument is connected in series with a synchronous signal amplifier, an integrator, an adjustable comparator and a phase inverter; both the two paths of the output terminals are connected with a synchronous detector which is connected with a low-pass filter, a regulating circuit, an A / D converter and a display; the regulating circuit is connected with a window comparator, a low-frequency signal generator, an audio signal generator and acoustic and optical alarm equipment; a signal processor of a global positioning system (GPS) is connected with a GPS time display and a central processing unit respectively; and the lightning alarming instrument and the A / D converter are connected with the central processing unit respectively.

A dual-fluid nozzle atomizing experiment device adapted to various atomizing media comprises a compressed air system, a water vapor system, a mixing system, a material liquid system and a circuit and control system for adjusting all control mechanisms. Ejection medium flow output by a compressed air and water vapor mixing system and material liquid output by the material liquid system are connected into an air channel inlet and a liquid channel inlet of a dual-fluid nozzle through pipes respectively. According to the atomizing experiment device, compressed air and water vapor mixed in any proportion can be selected as ejection media to be adapted to different types of dual-fluid nozzles. Multiple parameters of temperature, pressure and flow of the ejection media and the material liquid are adjusted respectively, the adjustment range is wide, and the spray quality and the material liquid applying proportion can be effectively controlled; the device is adapted to various atomizing media and experiment parameters, an experiment platform is provided for detecting and evaluating the atomizing effect for enterprises, and the purposes that technological parameters of the ejection media and the material liquid are controlled effectively, the spraying performance and stability are improved, and the flavoring and charging technological quality of tobacco is improved are achieved.

The method incldues the following steps: writing in and fetching real-time data, using shared internal memory mode as internal memory of real-time databank and ensuring data uniformity by write / read lock; picking up real-time data locus and to directly pick up fetching data from databank for those high-level application which wants to obtain locus within front 1s, synchronizing the real-time data, picking up the real-time data from each phasor measuring storage zone in sequence as per lot number corresponding to this moment by the databank, using long integer as indexing to quickly locate stored position of the rea-time data.

The invention provides a dynamic texture synthesis method based on the double-identification flow convolutional network. The method mainly comprises steps of design of an apparent flow in a texture model, design of a dynamic flow in the texture model, new texture generation, dynamic texture synthesis and texture style conversion. The method is characterized in that target identification and optical flow prediction are accomplished based on the pre-training convolutional network (ConvNet), after given dynamic state texture input, each appearance of input textures is packaged according to filter data responded by a target identification convolutional network, modeling of a dynamic state of the input textures is carried out according to filter data responded by an optical flow convolutional network. For generating new textures, optimization of a noise input sequence is required to match characteristics of each identification flow in the input textures, and appearance of one texture and a dynamic state of another texture are combined to generate a brand new dynamic texture. The method is advantaged in that a high quality sample matched with pixel-by-pixel appearance and time-varying conditions of the input textures can be generated.

The invention provides a power distribution network single-phase earth fault positioning device and method based on edge calculation. The device comprises: a positioning calculation module which is arranged in the automatic main station of the power distribution network and measuring points which are distributed in the whole network; each measuring point is provided with a measuring device; and each measuring device comprises a three-phase acquisition module installed on a power transmission line and a convergence module installed on a telegraph pole. The three-phase acquisition modules realize synchronization of each measurement point through Beidou / GPS time service. Current waveforms are synchronously collected in real time, wave recording data are stored locally, when a single-phase earth fault occurs, edge calculation is executed locally to process the fault waveforms, wave heads of the fault point waveforms are searched for, and the current waveforms of multiple cyclic waves before and after the fault and time information of the current waveforms are transmitted to a convergence module; and the convergence module wirelessly transmits the fault information data to a positioningcalculation module of a power distribution network automation master station, calculates to obtain a fused traveling wave velocity, and calculates a fault point position by adopting a double-end traveling wave distance measurement algorithm according to the fault information data.

One implementation involves a device receiving a stream of pixel events output by an event camera. The device derives an input image by accumulating pixel events for multiple event camera pixels. Thedevice generates a gaze characteristic using the derived input image as input to a neural network trained to determine the gaze characteristic. The neural network is configured in multiple stages. Thefirst stage of the neural network is configured to determine an initial gaze characteristic, e.g., an initial pupil center, using reduced resolution input(s). The second stage of the neural network is configured to determine adjustments to the initial gaze characteristic using location-focused input(s), e.g., using only a small input image centered around the initial pupil center. The determinations at each stage are thus efficiently made using relatively compact neural network configurations. The device tracks a gaze of the eye based on the gaze characteristic.

The invention discloses an infrared interferogram reconstruction method, wherein a laser interference signal and an infrared interference signal which are generated through light splitting by a Michelson interferometer exist. The method comprises the steps of simultaneously performing oversampling on the laser interference signal and the infrared interference signal by means of the same sampling rate; according to the requirement in an actual condition, setting the same interpolation factor for performing interpolation operation on two signals after oversampling; acquiring laser zero-crossing information from the laser interference signal after interpolation, and performing resampling on the infrared interference signal after interpolation according to the laser zero-crossing information, and finally obtaining an infrared interferogram with equal light path difference after reconstruction. According to the method of the invention, sampling and interpolation operation on the two signal are performed simultaneously, thereby ensuring time consistency and simplifying extraction to the zero-crossing information of the laser interference signal. The method according to the invention reduces data processing complexity.

The embodiment of the invention provides a processing method, server and system for a distributed timestamp. The processing method includes the steps that when a timestamp is signed and issued, a file summary matched with a file identification is sent to an opposite terminal server; the opposite terminal server obtains a first Hash timestamp; when the timestamp is identified, a to-be-verified file is processed, a second Hash timestamp is obtained, the second Hash timestamp is encrypted to obtain a second signature, and the second signature matched with the to-be-verified file is sent to the opposite terminal server; the opposite terminal server decodes and obtains the second Hash timestamp, and when the first Hash timestamp and the second Hash timestamp are same, verification is passed. By means of the processing method, the user requirements with the huge number can be met, the response efficiency of the system is improved through concurrent processing, and the processing method has the good expansibility; the time consistency of the distributed server is guaranteed through offset correction.

The invention provides a depth map processing method and a depth map processing device. The method comprises the steps: determining a measurement error according to a motion parameter during the collection of a first depth image frame; determining a depth difference value between adjacent image frames; determining credible pixels and non-credible pixels in the first depth image frame according tothe depth difference value and the measurement error, and determining a first smoothing factor corresponding to the credible pixels and a second smoothing factor corresponding to the non-credible pixels; and filtering the depth value corresponding to the credible pixel according to the first smoothing factor, and filtering the depth value corresponding to the non-credible pixel according to the second smoothing factor. Therefore, the measurement error corresponding to the depth value is adaptively determined based on the time dimension. The credible pixel and the non-credible pixel in the depth map are determined based on the measurement error, so that different error smoothing processing can be performed on different pixels. The time consistency of the depth value error after image framefiltering is ensured.

The invention provides a depth map processing method and device. The method comprises the following steps: obtaining a first depth image frame and a second depth image frame adjacent to the first depth image frame, determining a first content value of each first pixel and a second content value of a corresponding second pixel, and obtaining a content difference between the first content value andthe second content value, determining a credible pixel in the first depth image frame according to the content difference, determining a smoothing factor corresponding to the credible pixel, and performing filtering processing on the depth value of the credible pixel according to the smoothing factor and the depth value of the pixel corresponding to the credible pixel in the second depth image frame. Therefore, time-based consistency filtering is carried out on the depth value in the depth slow change area based on the change condition of the depth value. The depth value of the depth smooth area in the time dimension is smoother effectively, and the original high dynamic property of the depth fast change area is kept.

The invention discloses a MSR-based high dynamic range video generation method, and the method comprises the steps: firstly carrying out the scene switching detection of a continuous frame, and stopping the subsequent flicker detection if the scene switching happens between a current frame and a former frame; decomposing the image into a detail layer and a base layer based on an MSR algorithm, enhancing the detail layer of the bright area, globally expanding the base layer, fusing the two enhanced layers, and performing color correction according to the original image, the brightness of the original image and the brightness image obtained by fusion to generate a high dynamic range image; and finally, flicker detection is performed on the generated continuous high-dynamic-range images according to the logarithmic geometric mean difference between the brightness of the current frame and the brightness of the previous frame, and if a flicker phenomenon occurs, the brightness of the current frame is processed to obtain a flicker-removed high-dynamic-range video frame. According to the method, image details can be better expanded, so that the region with poor exposure is better expressed; and meanwhile, the time consistency of the video frames is ensured.

The embodiment of the invention discloses a method for positioning an address of a snapshot disk. The method comprises the steps: judging whether a queried snapshot record exists in a snapshot version table in a data disk or not when a request for looking up an address of the snapshot disk is detected; further judging whether a snapshot version number exists in the snapshot version table or not when the snapshot record exists in the snapshot version table; determining the address of the snapshot disk according to a snapshot meta address and snapshot version number corresponding address map table in the snapshot version table when the snapshot version number exists in the snapshot version table; determining a second snapshot version number meeting preset conditions in the snapshot version table when the condition that the snapshot version number does not exist in the snapshot version table is judged, and determining the address of the snapshot disk according to a snapshot meta address and second snapshot version number corresponding address map table. The efficiency of snapshot disk positioning is greatly increased, and the flexibility of snapshot is greatly improved. In addition, the embodiment of the invention further provides a corresponding implementation device, so that the method can have higher practicability. The device has corresponding advantages.

The invention provides an automatic parking method and system for a vehicle, and relates to the technical field of Internet of Vehicles automatic driving. The automatic parking method comprises the steps that the time of a vehicle is synchronized with the time of an edge cloud; acquiring driving environment video data, motion state information and time corresponding to the driving environment video data of the vehicle in real time; performing automatic parking according to the driving path planned by the edge cloud; when the vehicle receives an automatic parking stop instruction sent by the mobile terminal, automatic parking is stopped, wherein the automatic parking stop instruction is selectively sent out by a user of the vehicle according to driving environment video data, motion state information and time delay displayed on the mobile terminal, and the driving environment video data and the motion state information are obtained from the edge cloud by the mobile terminal; the time delay is obtained by the mobile terminal from the time corresponding to the driving environment video data acquired from the edge cloud. According to the automatic parking method, accurate time delay can be obtained in real time at the mobile terminal.

The invention discloses a full-configuration real-time simulation test method based on double-satellite hardware in a loop. The method comprises the following steps: step 1, two sets of satellites respectively carry real-time simulation test systems; step 2, data communication mechanisms of the two sets of test systems perform communication interaction by using a fixed-period mutual transmission mode, one set of test system is used as a time sequence main control end, the other set of test system is used as a controlled end, and the controlled end receives a rising edge or a falling edge of amain control end signal to complete time correction work in a triggering mode; and step 3, a software time supplementing mode is used to ensure the time consistency of the track data, so that the timebetween the two sets of test systems is synchronized. According to the invention, the communication protocol interaction can be carried out through two modes of RS422 serial port communication or UDPnetwork port communication under an engineering constraint condition, and the formation test task of the double-satellite formation is completed through a time synchronization method.

The invention discloses a method for unidirectional data synchronization between multiple data sources and relates to the technical field of data storage. Through carrying out mapping management on data between master and slave data sources, configuration management on unidirectional data synchronization between the master and slave data sources and distributed scheduling management on data transmission tasks, making transmission nodes carry out parallel transmission on the data and changing and collecting the data between the master and slave data sources, data synchronization is realized, and the problems of consistency and timeliness of the data between the master and slave data sources of the existing distributed data storage are solved. The method has the advantages of strong fault tolerance, simplicity in operation and data change timing consistency.

The invention discloses a write operation method of virtual machine block equipment and a snapshot creation method and device. The virtual machine block equipment at least comprises more than two sub-blocks. The write operation method comprises the following steps: when indication information indicating that a first sub-block creates a new snapshot from a first service node is received, suspendinga write operation for the first sub-block; inquiring whether the new snapshot of the second sub-block is created completely or not to the second service node so as to determine whether the snapshot of the current virtual machine block equipment is created completely or not; when it is determined that snapshot creation of the current virtual machine block device is completed, recovering a write operation for the first sub-block; wherein the virtual machine block device at least comprises more than two sub-blocks; wherein the second sub-block is a preset snapshot synchronization point, and thefirst sub-block is any sub-block which is not the snapshot synchronization point in the virtual machine block device and is performing a write operation. According to the method and the device, distributed snapshots of virtual machine block equipment segmented into a plurality of sub-blocks can meet the time consistency.

The invention relates to a control unit for a train running monitoring vehicle-mounted system. A two-out-of-three safety redundancy technique is adopted for controlling the output. The control unit comprises a three-mould collecting module and a three-mould master control unit connected with the three-mould collecting module, wherein the three-mould collecting module comprises a three-mould switching value collecting unit and a three-mould speed collecting unit; the three-mould master control unit comprises three input units which are communicated with each other; the three-mould master control unit is used for sending processed input information into the control unit; the three pieces of input information is voted by the control unit; after the two-out-of-three vote, a vote result is outputted. According to the technical scheme provided by the invention, the control unit in three moulds and the input units are identical to each other and the input signal sources of the input units are the same; the time consistency for collecting data of the three moulds is ensured; the correctness of the data information before three-mould calculation (data exchange before calculation and two-out-of-three vote) is ensured; the correctness of an outputted instruction is ensured.

The invention discloses a unified time sequence control and cooperative processing method for a radar system. The method includes the following steps that: a radar control system generates time sequence signals in different forms on the basis of a time sequence according to the processing requirements of a transmitting system and a signal processing system and transmits the time sequence signals;the signal processing system extracts and processes received echo signals according to a transmitting time sequence and a receiving time sequence which are transmitted by the control system, so as toeliminate invalid leakage signals; data interception is performed on the received processed echo signals according to system requirement design, and therefore, signals within a coverage range requiredby the system are reserved, and signals outside the coverage range required by the system are removed; and the signals within the coverage range required by the system are synthesized according to different distances. With the method of the invention adopted, the extraction of effective echo signals is ensured, additional resource consumption that does not contribute to system requirements is reduced, and the problem of excessive resource consumption of traditional schemes is solved, and system operation efficiency is effectively improved.

The embodiment of the invention discloses an Internet of Things message issuing method and device, electronic equipment and a storage medium. The Internet of Things message issuing method can comprisethe following steps of sending a request message for requesting synchronization; after the request message is sent, a receiving window being opened; receiving a response message returned for the request message through the receiving window; determining whether the time of the terminal device and the time of the gateway device are synchronous according to the receiving condition information of theresponse message; if the time of the terminal device and the time of the gateway device are synchronized, adding the terminal device and the gateway device into a multicast group; and receiving a downlink multicast message sent by the management equipment based on the multicast group. Therefore, batch control of a plurality of devices can be realized through one or a few downlink data.

The invention discloses a method of dynamic soft tissue motion prediction for an HIFU treatment system. The method comprises the following steps: firstly, preprocessing an HIFU image sequence; constructing a prediction mapping relationship between the image sequence and a label sequence; constructing a data set and carrying out online data enhancement on a training set; building and training a dynamic soft tissue motion prediction model of the HIFU treatment system and acquiring an optimal model through an evaluation index optimization network; and after the optimal model is obtained, inputting the test set into the network, and carrying out dynamic soft tissue motion prediction and performance evaluation of a medical image sequence. A combination mode of combining the stacked convolutionlong-short-term memory network with a time wrapper is adopted to provide basic data conditions for a network model to process a medical image sequence, space-time information of an HIFU image sequencecan be fully utilized, and motion prediction with an accurate contour can be carried out on dynamic soft tissue of an HIFU treatment system.

The invention provides a depth map processing method and device, and the method comprises the steps: obtaining a first depth image frame and a second depth image frame adjacent to the first depth image frame; determining a first content value of each first pixel and a second content value of the corresponding second pixel, and obtaining a content difference between the first content value and thesecond content value; determining a credible pixel in the first depth image frame according to the content difference, and determining the area of a region where the credible pixel is located; and when the area of the region meets a preset condition, adjusting an acquisition interval of the depth image frame according to the area of the region, and obtaining a next to-be-processed depth image frame according to the acquisition interval. Therefore, the sampling interval of the image frame is determined based on the area of the area where the depth value changes gently, and the calculation amount of time consistency filtering is reduced.

The invention discloses a hybrid power transient state measurement control test system. The system comprises a test room, a large bottom plate and an electric dynamometer; the large bottom plate is arranged at the bottom of the test room; a shock absorber is arranged between the test room and the large bottom plate; the electric dynamometer, an analog electronic throttle, an analog turbocharged intercooler and a hybrid power installation bracket are arranged on the large bottom plate; hybrid power is arranged on the hybrid power installation bracket; the installation bracket is provided with an adjustment screw rod; the adjustment screw rod is fixed on an adjustment seat through an adjustment frame; a shaft coupling, a torque sensor, an elastic shaft coupling and a transmission shaft are sequentially connected with the right end of the electric dynamometer; and a test system is arranged on one side of the large bottom plate. The hybrid power transient state measurement control test system has the advantages that an overall hybrid power vehicle serves as a control target, the verification of relative properties, such as the dynamic property, the emission performance and the economical efficiency of the power assembly of a passenger vehicle and a commercial vehicle are realized, and the verification of an overall vehicle strategy related to the power assembly and the verification test of related components can be performed.

The invention discloses a real-time computing method and system based on multi-device out-of-order data. The real-time computing method based on the multi-device out-of-order data comprises the stepsthat a real-time database collects the real-time data from multiple devices; the real-time database is read in a preset reading period, and for each reading period, the latest real-time data of multiple devices in the time range of the current reading period are obtained; and operation is carried out on the latest real-time data according to a preset algorithm to obtain an operation result. According to the invention, by setting the reading period and reading the real-time database in a reading period, obtaining the latest real-time data collected from a plurality of devices within a current reading period, and taking the acquired latest real-time data of the multiple devices as the same timestamp data to be calculated, the problem of real-time data disorder is solved, the time consistencyof the data disorder of the multiple devices is kept, and the latest real-time data of the same timestamp can be simply and conveniently calculated according to the preset algorithm to obtain the calculation result.

The invention discloses a plane shunting system with an air interface timing function and a timing method. The system comprises a district station, n locomotive stations and n flat regulation handsets. The district station is connected with the n locomotive stations and the n flat regulation handsets through a wireless air interface network. Remote software is installed in a PC. The PC and the district station are connected through a TCP / IP network. The air interface timing is carried out through an air interface network to exchange messages in the form of signaling. The system is convenient and simple to implement, can cover all pieces of equipment in the air interface network, and saves the complicated process of time configuration for each piece of equipment, thereby being simple, convenient and quick; the time consistency of the equipment in the system is ensured by the timing method; the method adopts automatic timing, does not need human interference and greatly saves the manualtime; and the time consistency of operation between each piece of equipment in the system is ensured.

The invention discloses a method for testing a periodic running time of a hardware board card, which relates to the optical field, and comprises the following steps that: step 1: a signal line of a high-precision oscillograph after the time check is connected onto a test point (TP) of a test board card; step 2: a test code is inserted into a periodic running program of the test board card; step 3: a periodic running time of the periodic running program is determined according to a waveform of the test code displayed on the high-precision oscillograph; and step 4: the periodic running time obtained by the high-precision oscillograph is compared with a designed periodic running time to obtain an error value between the real periodic running time of the periodic running program of the testedboard card and the designed periodic running time. The high-precision oscillograph is used for testing the periodic running time of the board card, so a clock source which is used for the test is independent from a clock source of the board card, the problem of the inaccurate test periodic running time caused by the inaccuracy of a system clock can be avoided, and the time consistence of the entire device can be guaranteed.

The invention is applicable to the technical field of displays, and provides an aviation different-speed image motion compensation circuit, which comprises a main time sequence generator, used for generate driving time sequence required by a CCD; a different-speed image motion time sequence controller, used for forwarding the driving time sequence generated by the main time sequence generator andgenerating an image motion compensation driving time sequence during different-speed image motion compensation; and a driving circuit used for amplifying and translating the driving time sequence output by the different-speed image motion time sequence controller so as to drive the CCD. The main time sequence generator is added to the aviation different-speed image motion compensation circuit, andthe main time sequence generator uniformly calculates various compensation parameters during different-speed image motion compensation, so that the time sequence consistency during different-speed image motion compensation is effectively ensured, and the effect of different-speed image motion compensation is better improved.

The invention discloses a tunnel active source seismic wave wireless acquisition method, system and terminal and relates to the technical field of tunnel engineering. The implementation process is as follows, an upper computer is in communication connection with a wireless access master station, and the wireless access master station is in wireless connection with a plurality of wireless acquisition terminals; starting up is carried out, and sampling parameters are set through the upper computer; sampling is started, the master station carries out AD sampling and issues a sampling starting instruction to the terminal, and the terminal starts AD sampling; the main station reads an AD sampling result, and when the result exceeds a set threshold value, the main station transmits a trigger completion instruction to the terminal and the upper computer at the same time; the upper computer starts timing after receiving a trigger completion instruction, and the terminal completes sampling after sampling time is set; and the upper computer issues a data uploading instruction to the accessed terminal, the terminal uploads the stored AD sampling data to the upper computer, and the upper computer stores and analyzes the received data and displays the waveform of each wireless acquisition terminal and each channel. The sampling data of the method has characteristics of high precision and traceability.

The invention discloses a verification method for an algorithm module in a chip, and the method comprises the steps: carrying out the analysis of an algorithm module verification request, and obtaining a to-be-verified target algorithm module; calling a target SystemC reference model corresponding to the target algorithm module; obtaining each test case corresponding to the target algorithm module through a preset input agent, and respectively inputting each test case into the target SystemC reference model and the target algorithm module according to a preset time sequence; utilizing an output agent to respectively obtain each level of output of the target SystemC reference model and the target algorithm module; and performing corresponding comparison on each level of output to obtain a verification result of the target algorithm module. According to the method, the time sequence consistency of the target SystemC reference model and the target algorithm module is ensured, the workload is greatly reduced, and the verification efficiency is improved. The invention further discloses a device, equipment and a storage medium, which have corresponding technical effects.