123results about How to "Improve real-time processing performance" patented technology

In a microprocessor that internally has a microprocessor specific secret key, a key management unit is provided to carry out a key registration for reading out from an external memory a distribution key that is obtained in advance by encrypting the instruction key by using a public key corresponding to the secret key, decrypting the distribution key by using the secret key to obtain the instruction key, and registering the instruction key in correspondence to a specific program identifier for identifying the program into a key table, and to notify a completion of the key registration to the processor core asynchronously by interruption when the key registration is completed.

The invention belongs to the field of computer vision and pattern recognition and discloses a moving target tracking method based on an improved multi-example learning algorithm. Firstly, a random measurement matrix is designed according to the compression perception theory. Then a multi-example learning algorithm is used to sample an example in a current tracking result small neighborhood to form a positive package, and at the same time, sampling an example is carried out in a large neighborhood ring to obtain a negative package. For each example, the characteristic of a character target is extracted at an image surface, and the random measurement matrix is utilized to carry out dimensionality reduction on the characteristic. According to the extracted example characteristic, online learning weak classifiers are utilized, and weak classifiers with strong discrimination ability are selected from a weak classification pool to form a strong classifier. Finally, when a new target position is tracked, according to a similarity score of the current tracking result and a target template, the online adaptive adjustment of classifier update degree parameters is carried out. According to the method, a problem that a tracking result in the existing algorithm is easily affected by an illumination change, an attitude change, the interference of a complex background, target fast motion and the like is solved.

The invention discloses a message reliable processing guarantee method of a real-time flow calculating frame based on Storm. The method comprises the steps: 1, in a preprocessing phase, initialization work is carried out on an environment; 2, data which have already been sent and are in a calculating state are tracked in a cluster calculation process; 3, when an emission task monitors a signal that the message processing is successful, tracking information of all sub-tuples which belong to the task in a cache region are emptied; 4, when the emission task monitors a signal that the message processing is unsuccessful, the position of the task producing the processing failure and data to be recovered are located; 5, a message recovery program is constructed according to the tracking information and xml files, the data to be recovered are read from the cache region, and the message recovery program is executed; 6, the cache region is emptied, and a memory space is released. According to the method, when a message is recovered, large-scale repetitive computation existing in a complex topology business is avoided, the calculated amount of message recovery is effectively reduced, the performance of processing data under the whole business scene is promoted, and the low delay requirement of real-time processing can be ensured.

The invention relates to a method for realizing a visual SLAM semantic mapping function based on a cavity convolutional deep neural network. The method comprises the following steps of (1) using an embedded development processor to obtain the color information and the depth information of the current environment via a RGB-D camera; (2) obtaining a feature point matching pair through the collectedimage, carrying out pose estimation, and obtaining scene space point cloud data; (3) carrying out pixel-level semantic segmentation on the image by utilizing deep learning, and enabling spatial pointsto have semantic annotation information through mapping of an image coordinate system and a world coordinate system; (4) eliminating the errors caused by optimized semantic segmentation through manifold clustering; and (5) performing semantic mapping, and splicing the spatial point clouds to obtain a point cloud semantic map composed of dense discrete points. The invention also relates to a system for realizing the visual SLAM semantic mapping function based on the cavity convolutional deep neural network. With the adoption of the method and the system, the spatial network map has higher-level semantic information and better meets the use requirements in the real-time mapping process.

The invention discloses a heterogeneous multi-core infrared image processing system and method. The method mainly includes: 1, utilizing programmable logic of a Zynq to realize a driving schedule of a detector to complete collection and caching of an infrared image; 2, utilizing a CPU0 (central processing unit 0) of the Zynq to operate a Linux operating system to write a Qt-based display and control interface; 3, utilizing a CPU1 of the Zynq to respond to a serial port command in real time, and setting a system parameter and a video driving chip online; 4, utilizing the programmable logic to realize algorithmic hardware acceleration and a driving schedule of a video chip. The heterogeneous multi-core infrared image processing system and method has the advantages that a structure of ARM+FPGA in an all programmable platform Zynq chip is utilized to develop the heterogeneous multi-core infrared image processing system, design difficulty of hardware is remarkably reduced, a system bandwidth chokepoint brought by interconnection of conventional framework chip levels is removed, simplicity and flexibility in design and integration of a user IP (internet protocol) are realized, and high universality is realized.

The invention is concerned with the navigation system and the emendation method of the navigation orientation data, in order to conquer the shortage of the existing technology, provide a emendation method that can emendate the navigation orientation data accurately especially fit for farmland busywork. The invention is: the map registration provides the more accurate electronic map to the system, compares the orientation data point of every sensor and the known skyway that prearranges to be track, and then processes general estimation, works out emendation data point with higher reliability. The invention is consists of four processes namely the map registration, the sensor data prearrangement, the emendation value decision-making, the new orientation data point creating.

A automatic fire unit based on comprehensive vision sensor comprising microprocessors, video frequency sensors for the surveillance, communication modules used to communicate with the outside world and firefighting nozzles, the firefighting nozzle is connected to drives, the video frequency sensor is a multi-dimensional visual sensors, dispose and analysis images, make a comprehensive judgment through the simple pattern classification, the flame area characteristics judgment, the flame physical characteristics judgment, flame color judgment, the law of flame flashing judgment, flame movement characteristics judgment and judgment of the fire intensity, automatically adjust the nozzles according to flames position and spray extinguishing agent assigned in the vicinity of the fire source. The invention can effectively reduce the false fire alarm rate, inspect and locate accurately, with good real-time performance and automatically control the nozzle spray fire-extinguishing agent to the flames found in the shortest time, the achieved detection and control is simple.

The invention provides signal processing equipment of a PET detector based on a neural network localizer. The equipment integrates and realizes procedures for processing a series of signals from a photoelectric switching signal outputted by a detector to the action position coordinate of the gamma ray counted by a neural network in real time. The equipment comprises a compression read processing circuit to an array pixel signal outputted by a photoelectric switching device, an analog-digital switching circuit, an energy identifying and timing circuit, a proper time judging circuit, a base line recovery circuit, a signal peak value estimating circuit, a neural network real-time computing circuit, a USB interface circuit, etc. Due to the realization of the technology for processing various digital nuclear signals based on FPCA and especially the real-time computing of the neural network, the instantaneity and the integration level of the system are greatly improved. The invention has compact structure, complete performance, strong on-line data processing performance, and can be conveniently assembled into an integral detector module with the PET detector, wherein the detector module is important intermediate equipment for researching and developing novel PET imaging equipment based on neural network positioning.

The invention provides an identification method for a radar disoperative target based on a mixed model, which is used for solving the problem that the disoperative target is low in identification rate. The method comprises the following steps: establishing a standard body model library of a refined scattering point model; structurally decomposing the disoperative target according to the standard body model library to generate a first scattering point model; shielding the first scattering point model to obtain an effective scattering point model; calculating RCS intensity for the effective scattering point to obtain intensity information and combining the intensity information to generate a scattering point matrix; adding a statistic characteristic into the scattering point matrix to obtain a second scattering point model of the disoperative target containing coordinate information and the RCS intensity information; carrying out multi-scattering point radar return simulation on the second scattering point model to establish a high resolution one-dimensional range profile template library; and identifying the tested high-resolution one-dimensional range profile by adopting a K near neighbor classifier by virtue of the high resolution one-dimensional range profile template library. According to the method provided by the invention, the target identification performance of the radar system can be improved.

The invention provides a joint inversion method of dielectric constant and concealed target parameters of homogeneous background media, which is realized by the following ways: firstly setting an estimation interval of a dielectric constant, then computing each estimation value in the interval to generate dictionary matrixes, then arranging the dictionary matrixes in rows to generate a joint dictionary matrix, carrying out optimization solution by utilizing imaging algorithm based on compressed sensing on the basis of the joint dictionary matrix to obtain a joint target matrix, then segmenting and extracting the joint target matrix and simultaneously obtaining the dielectric constant and the imaging results of the detection regions of the background media, thus improving the real-time detecting and imaging capabilities of the ground penetrating radars to a greater extent.

The invention discloses an unmanned aerial vehicle (UAV) data link universal access system based on communication satellites, which is applied in a communication satellite control and management center. The system comprises a multithread multitask processing module, a network distribution module, and an IPC data exchange module, wherein the network distribution module is used for configuring ports and IP addresses for data senders and data receivers in an offline manner; the multithread multitask processing module is used for creating a master control process after data is received from the data senders, the master control process comprises a plurality of threads, the received data is split through the threads, and the split data is repacked into data frame formats required by the data receivers; and the IPC data exchange module is used for scheduling the threads during processes. The system provided by the invention has the advantages that UAV ground stations needn't respectively construct relevant protocol conversion modules, so that the link access efficiency and the UAV interoperability are improved, and communication satellite communication transmission protocol requirements are satisfied.

The invention discloses an embedded high-space and high-spectral resolution video acquisition system. A video acquisition card of the system conveys the acquired RGB and gray scale videos to a memory, and a processor decodes the videos to separate RGB image signals and gray scale image signals and synchronously aligns and corrects the RGB image signals and the gray scale image signals; a principal component analysis method is used for reducing the spectral information dimension of pixel points so as to simplify the calculated amount; then a spectral propagation algorithm of bilateral filtering is executed on the processor in a multi-thread parallel mode so as to obtain spectral information of all pixel points of video frames; high-resolution spectral information is recovered by the principal component analysis method; data is stored in the memory, and acquisition, processing and analysis processes are displayed by display equipment; and finally, spectral data in the memory is transmitted to a remote host through the network. The system disclosed by the invention realizes integration of acquisition, processing, analysis and display processes of the videos, has the characteristics of miniaturization and integration, and is convenient to carry and use.

The invention relates to a pulse wave motion artifact removal method for parameter adaptive optimization VMD, and belongs to the technical field of biomedical signal acquisition and processing. The method comprises the following steps of by optimizing an objective function of variational mode decomposition, evaluating the variational mode decomposition result by adopting a correlation coefficientwith the fusion of the intrinsic mode correlation degree and the central frequency dispersion degree; carrying out parameter adaptive optimization variational mode decomposition on a pulse wave signalto obtain an intrinsic mode [uk] of the pulse wave, and designing a denoising rule according to the relative energy of frequency bands corresponding to different components of the signal to remove the motion artifacts of the pulse wave. Compared with the prior art, the method has the advantages that the quality of pulse wave variational mode decomposition is improved, the effect of removing the motion artifacts is improved, and the data complexity in the process of removing the motion artifacts is simplified.

The invention relates to a real-time data collaboration method in an industrial internet of things. The method provides a basic data coordination mechanism for application and implementation of industrial software in an industrial production site. The data collaboration method, through a pre-deployment mode, deploys data to nodes about to be used by the data before the data is used; meanwhile, themethod does not depend on a center service node, can realize data collaboration and information flow in a high-reliability network of a production environment better, improves data obtaining time andimproves data real-time processing performance; and meanwhile, the method improves data collaboration mode and flow path, and can improve data collaboration and flow reliability and performance.

The invention discloses an image tracking system comprising a video image acquisition module, an FPGA and a DSP, wherein the video image acquisition module is used for shooting video images and outputting digital image data. The FPGA is used for storing a present frame of digital image data which is output by the video image acquisition module through a digital signal output end thereof in to a first storage area of a storage via a data storage port of the FPGA. Digital image data of a previous fame stored in a second storage area of the storage is read through a data read port of the FPGA, and after the digital image data is subjected to gray scale transformation and median filtering, the data is output through a digital signal output end of the FPGA. The DSP unit is used for carrying out target searching and tracking according to the digital image data output by the FPGA. Through the utilization of the image tracking system and the image data processing method of the invention, capability of performing real time processing on the image data can be improved.

Provided are an apparatus and method for executing a telecommand on a geostationary satellite, and an apparatus and method for verifying a telecommand execution status on a geostationary satellite ground control system. When a telecommand on a satellite is executed, the satellite generates command execution verification words for the executed telecommand and adds the generated command execution verification words to a telemetry transfer frame, and thus a satellite ground control system can easily verify the telecommand execution. In addition, when telecommand execution result is verified by a satellite ground control system, since an execution result of a time-tag telecommand can be verified without a time delay through command execution verification words provided by a satellite, and a telecommand image DB, a telecommand verifier DB, and a telecommand verifier provided by the satellite ground control system, a series of processes from transmission to verification of a telecommand, in particular a time-tag telecommand, can be automatically processed in real-time without intervention of an operator.

The invention discloses a sphere decoding detection method and device. The method comprises the steps that a received signal is preprocessed, and a signal approximation estimation value Xpre of the received signal is acquired; according to the Xpre, the initial square radius D2 of sphere decoding detection is derived; according to the current signal to noise ratio of the received signal, the constellation space size l is determined; according to depth first and sphere constraint rules, and according to the constellation space size l and the initial square radius D2, a searching path is searched; nodes passed by the searching path are located in a sphere whose radius is the initial square radius; after the searching path is searched, and when the sum of the local euclidean distances of the searched searching path is less than the current square radius, the square radius is updated; in a multidimensional sphere, the searching path is searched again, wherein the received signal is the sphere center, and the updated square radius is the radius; and until the searching path cannot be searched, a candidate signal point corresponding to the searching path which is latest saved is determined as the optimal signal estimation point.

The invention provides a real-time data processing device, method and system based on the Internet of Vehicles. \The device comprises: the application server which is used for processing service requests of the APP side and the Web side; a terminal access service which is used for receiving vehicle information and external information uploaded by a vehicle-mounted terminal, converting the information into the preset message and storing the preset message; the message middleware which is used for pushing the vehicle information and the external information to the streaming computing frameworkand the service processing module; a flow calculation framework for calculating a speed of a moving vehicle based on position data, time data, and road conditions, and comparing the calculated speed of the moving vehicle with a preset vehicle speed to generate a corresponding control command,; the service processing module which is used for tracking and monitoring the moving vehicle in real time according to the received vehicle information and external information; and the data storage layer which is used for storing data. The technical effect of improving the real-time processing capabilityand intelligence can be achieved.

The invention provides an exception handling method and structure tolerant of missing cache and capable of emptying an assembly line quickly. The exception handling method includes the steps that firstly, a false hit state is defined, and the access result of the cache by invalid commands emptied by the assembly line is marked; correct command codes or data words are not guaranteed by the cache under the false hit state, and the corresponding cache information of access addresses is output directly; an assembly line emptying signal is directly acted on the command cache in the command taking operation in the false hit state to enable the false hit state to be selected when the invalid commands have access to the cache; when exception handling begins, data cache is selected and controlled through the assembly line emptying signal and the or logic of a storage access level invalid marking signal; when an exception occurs, the cache enters the false hit state to enable the assembly line to advance continuously. The invention further discloses the structure applying the exception handling method. By means of the exception handling method and structure tolerant of missing the cache and capable of emptying the assembly line quickly, the exception handling speed can be increased, and the real-time performance of the system is improved.

The invention provides a Monte-Carlo location algorithm based on a self-adaptation iteration volume particle filter, and belongs to the technical field of robot location methods. According to the Monte-Carlo location algorithm based on the self-adaptation iteration volume particle filter provided by the invention, the improvement is carried out on the basis of a CMCL (Monte-Carlo location of volume particle filter) algorithm, and the new location algorithm is provided aiming at the problems of high computing capacity and poorer real-time processing capacity existed in a current CMCL algorithm.According to the Monte-Carlo location algorithm based on the self-adaptation iteration volume particle filter provided by the invention, a Gauss-Newton iteration volume Kalman filter is used for generating the volume particle filter with significant proposal distribution; a Kullback-Leibler distance criterion is utilized for enhancing CPF (Continuation Power Flow), so that the particle quantity is self-adaptively selected. According to the Monte-Carlo location algorithm based on the self-adaptation iteration volume particle filter provided by the invention, the limitation of high-order truncation errors of the standard volume particle filter can be effectively avoided, the location error and the algorithm calculation quantity are reduced, and the processing capacity is improved.

The invention discloses an SINS large-azimuth misalignment angle initial alignment method based on a self-adaptation UPF. The method includes the following steps that 1, an SINS non-linear error model is built; a misalignment angle between an ideal navigation coordinate system and a computed navigation coordinate system is presented by means of an Euler platform error angle, and the corresponding SINS non-linear error model is built according to the rotation sequence of the error angle. The method reduces influences caused by system simplification and noise statistic feature uncertainty on a system to some extent. Meanwhile, the UPF middle particle degradation phenomenon is well eliminated. Initial alignment precision of a strap-down inertial navigation system is improved, and alignment time is shortened. The method is a software method, system hardware does not need to be modified, and accordingly the method is convenient and feasible to practically implement.

The invention discloses a real-time detection and tracking system and method for a point target in a satellite sequence image. The system comprises an image data receiving module, an image frame header detection module, a single-frame image target detection module and a sequence image association detection module, wherein the image data receiving module receives a data package of a preset N-byte length from a satellite infrared image generation system each time and caches the data package in a ring buffer region; the image frame header detection module performs position detection on an initial position and an end position of the image in the ring buffer region and caches frame image data in an image cache region; the single-frame image target detection module is used for processing the image data of the image cache region to obtain suspected target information; and the sequence image association detection module is used for associating suspected target results of multi-frame image detection in time and space, and removing noise points to obtain real target information. According to the system and the method, the module division is optimized, so that the integrity, timeliness and reliability of data processing are ensured.

The invention discloses a high-resolution wide-swath onboard SAR (synthetic aperture radar) real-time motion compensation algorithm which mainly includes an envelop compensation portion and a phase compensation portion. A traditional phase compensation algorithm is improved, and phase compensation can be finished before envelop compensation. Envelop compensation approximate treatment needs to divide distance sub-bands for echo signals after distance compression, so that motion error space varying is reduced. Therefore, interpolation operation of a traditional envelop compensation algorithm is avoided, and envelop compensation is finished by directly multiplying distance frequency domain with linear phase. Finally, distance sub-band images are spliced to obtain an imaging result of a whole swath. By the aid of an improved phase compensation algorithm, signal phase precision is ensured, so that the interpolation operation with large calculating amount can be more effectively avoided by rapid envelop compensation based on the distance sub-bands, and the real-time treatment performance of the algorithm is improved without affecting resolution.

The invention belongs to the field of radar signal processing, and particularly relates to a parallel radar pulse compression method which uses a real-time operating system for real-time scheduling. The parallel radar pulse compression method includes the steps that CPU affinity API functions set in a linux kernel are used for binding the progress for achieving a pulse compression algorithm to a certain kernel of a multi-kernel CPU, so that the kernel of the multi-kernel CPU only carries out the pulse compression algorithm, based on kernel binding, the operating system uses an improved real-time scheduling mechanism and ensures real-time processing of the pulse compression process without being affected by factors such as system interrupt, external interrupt and other progress operation, and therefore real-time performance of the system is greatly improved. The method effectively improves the implementation efficiency of the pulse compression algorithm, and radar signals can be processed in real time.

The present invention provides a video object tracking system and method. The system comprises: an image collection module configured to shoot video images; an image transmission module configured toperform input and output of images in the video shot by the image collection module; an FPGA module configured to perform real-time processing of the shooting video images transmitted by the image transmission module and output processed video image data; and a DSP module configured to receive the video image data processed by the FPGA module, perform operation, determine a tracking object and return the related image data after operation to the FPGA module. The FPGA module is configured to perform analyze of transmitted and collected video data signals and send an operation result to the DSPmodule to perform reoperation of an object to determine a concrete tracking path of the object so that the real-time processing capacity of a high-speed moving object in a high-resolution video imageis improved and an operation speed is improved by above 95% compared to that of like products.

The invention discloses a cooperative-target-based real-time global search method for high-speed vision measurement and relates to a marking-point search method. By using the real-time global search method, the problems that in the high-speed vision measurement, the global search process is overlong in time consumption and is sensitive to the resolution of a camera, and the updating rate of a system is limited by the moving speed of the target are solved. In a data preprocessing module, parallelized data are outputted to a data buffering control module and a marking-point real-time detection module simultaneously. After the marking-point real-time detection module detects a marking-point pixel, subsequent related data of marking points are sent to an operating module which is in a multi-production-line structure; and after the transmission of a frame image is ended, coordinates of all the marking points can be outputted, and mass-center data of the marking points are obtained by calculation and is stored into a corresponding cache. The real-time global search method disclosed by the invention has the advantages that the global search process is short in time consumption, good in robustness and insensitive to the resolution to the camera and the moving speed of the target. The real-time global search method is applicable to the image processing field and the vision-measuring field.

The invention provides a method for enhancing the visibility of an image, which comprises the following steps of: obtaining information of at least two images of a scene; aligning the image information to a pre-established scene three-dimensional model to obtain a template image corresponding to the image information and depth information of the template image; estimating weather condition parameters when obtaining the image information of the scene according to the template image and the depth information of the template image; correcting the depth information of the template image according to the weather condition parameters to obtain the depth information of the scene image; and carrying out image enhancement on the depth information of the template image according to the depth information of the scene image. Through combining the scene three-dimensional model, the invention is used for effectively enhancing the visibility of a dynamically collected image with deficiency of textures and monotonous color, thereby ensuring that the collected image can better serve various applications.

The invention discloses a freshness sensitive big data summary information maintenance and aggregate value query method. The method comprises the steps that (1), a time tracker is built for time object data of each time object; (2), time object data to be written in are mapped to the corresponding time trackers according to the time objects, and then the trackers divide the corresponding time object data into multiple time stages and set the error parameters of each time stage; (3), the trackers sample the time object data within the time stage according to the error parameters of each time stage and store the time object data into sample sets corresponding to the time trackers. In the query process, firstly, the corresponding time trackers are positioned according to the keys of the time objects, then the trackers query the time stages in the time trackers according to the query time information, and a query value is returned according to the samples corresponding to the found time stages. The method can effectively manage and query the time object data and support a higher level of subject-oriented computing application.

The invention discloses a TDOA (time difference of arrival) positioning method, device and system based on time difference calculation. The method includes the steps: measuring and receiving signals transmitted by the same radiation source by receiving nodes of a plurality of measuring stations; processing the signals to remove high-frequency section spectrum line data of the signals, and transmitting the processed signals to a control center; receiving signals processed by different measuring stations by the control center; screening low-frequency section effective spectrum line data of the signals to remove spectrum line data with low contribution degree for time difference calculation; cutting down spectrum line amplitudes with low contribution degree for time difference calculation inthe signals processed by a discretized weighting function to obtain a cross-power spectrum density function, and calculating time difference to perform TDOA positioning. According to the method, noisesensitivity is reduced, the resolution ratio of time delay meets engineering requirements, transmitted data are suitably compressed, and real-time processing of the control center is accelerated.

The invention discloses an angle-based wandering track detection method in video surveillance, which comprises the following steps of: 1, performing moving target detection on each acquired frame of image; 2, after a target is detected, determining positions A and C of the moving target in two initial frames, and connecting the target positions n the two initial frames to generate a radial AC; 3, similarly, taking the initial point A of the moving target in the initial frame as a starting point, and calculating an included angle alpha between a radial AD passing each target position D and the radial AC; 4, determining the type of a track according to the changing regularity of the included angle alpha, judging whether the track is a wandering track or not, and setting a wandering threshold with an initial value 0; and 5, judging whether the wandering threshold exceeds the set threshold or not, giving a wandering alarm if the wandering threshold exceeds the set threshold, otherwise returning to the step 1 to reacquire a video sequence. The method ensures high accuracy and high applicability.