88results about How to "Take advantage of computing power" patented technology

The invention discloses an output method, system and server of recommendation information. The method comprises the following steps of: extracting user action data in a predetermined statistic period from a user action database; classifying the user action data according to commodity identifications to obtain a time sequence about interestingness of each commodity in the statistic period; computing a purchase peak probability of the commodity according to the time sequence about interestingness; when receiving a command of outputting the recommendation information, sorting the purchase peak probability according to an order from high to low, and outputting the recommendation information of the commodity according to a sorting result. The method provided by the invention automatically computes the purchase peak probability of the commodity on time dimension according to the user action data, thereby improving the recommendation accuracy of a recommendation system and reducing the transmission quantity of unnecessary data in the network; since the recommendation information is automatically modified and output by a server, and the computation capability of the server is sufficiently used, the information input quantity and computation resource of the recommendation system can be reduced when the recommendation information is adjusted according to the purchase peak probability.

The invention discloses a local data updating method based on an erasure code cluster storage system, which comprises the following steps: receiving a data D0 updating request sent by a user; updating a data D0 into a data D0'; sending the data D0' to a data node to be updated; reading the data D0' from a local disc to a memory by the data node to be updated; calculating a provisional data block set needed by m redundant nodes in the memory of the data node to be updated; writing the data D0' into the local disc by the data node to be updated; sending m provisional data blocks onto the m redundant nodes by the data node to be updated through the network at the same time; reading an original redundant data Pj by the j redundant node from the local disc, and calculating to obtain a new check data Pj'; and writing the check data block Pj' obtained through calculating into the local disc by the redundant node. Due to the adoption of the local data updating method based on the erasure code cluster storage system, the consumption of the bandwidth can be reduced effectively, the load pressure of the update server can be decreased, and the data updating performance can be improved.

The invention discloses a super-high-resolution three dimensional reconstruction method based on a global variation technology. The method comprises the following steps: low-resolution three dimensional geometrical model and a high-resolution surface normal graph of a scene to be constructed is obtained; a global optimization energy function is constructed and obtained according to the difference between the low-resolution geometrical model and the high-resolution normal graph; the minimum value of the global optimization energy function is solved, a depth value of each pixel point is obtained, and three dimensional reconstruction is carried out according to the depth values of all the pixel points. The super-high-resolution three dimensional reconstruction method based on non-local total variation regular constraint condition lowers dependence of a final reconstruction result on an original result effectively and raises the three dimensional reconstruction precision greatly.

The invention relates to a self-organized model constructing method of a cloud resource sharing peer-to-peer network, a distributed peer-to-peer network for short. The topology structure of the distributed peer-to-peer network is constructed through adoption of a space division concept. Physically closed cloud noses are self-organized. A reachable routing table is extended. The cloud peer-to-peer network which takes data as a core and is equipped with performances such as low delay and extensibility is realized. According to the method, in a large scale cloud resource distributed sharing environment, cloud node information is managed in real time, rapid resource positioning is carried out to the computing resource demand of each user, and an inquiring result is returned. Compared with the prior art, the method has the advantages that the method has clear average cloud resource inquiring delay advantage, the computing capabilities of the nodes are utilized fully, the actual performance of the cloud computing application is satisfied, and the method has wide application prospects in cloud resource sharing aspects.

The invention provides a personalized recommendation method and system for friends and applications. The method comprises the steps that A, a recommendation score S is calculated; B, recommended content is displayed according to the recommendation score; C, a recommendation weight vector is adjusted according to feedback information about the recommended content from a user; D, the recommended content is updated according to the adjusted user recommendation weight vector. The personalized recommendation method and system have the advantages that the application range is wider, recommendation results are more accurate, the calculating speed is higher, calculation at idle time is achieved, a blacklist mechanism is adopted, the accuracy of information recommendation can be effectively improved, calculation cost is saved, and a good recommendation effect is achieved.

The invention provides a multifunctional universal remote control system based on a mobile phone platform. The multifunctional universal remote control system based on the mobile phone platform comprises the mobile phone platform and a signal conversion module, wherein the mobile phone platform monitors instruction information of users, converts the instruction information into American standard code for information interchange (ASCII) code control commands according to a mapping table and then sends the ASCII code control commands to the signal conversion module, the signal conversion module receives the ASCII code control commands sent by the mobile phone platform and transmits the ASCII code control commands to a control single chip microcomputer, the control single chip microcomputer finds infrared signal binary codes corresponding to the ASCII code control commands according to the mapping table and outputs the infrared signal binary codes to an infrared communication module, and the infrared communication module sends the input signals in an infrared light signal form and realizes the household appliance control. The multifunctional universal remote control system has the advantages that the household appliance remote control function can be added to a mobile phone without changing hardware circuits of the household appliances and the mobile phone, the operation capability and good interaction performance of the mobile phone are sufficiently utilized, and the infrared control on the household appliances is realized through the voice identification technology or through mobile phone press keys.

The invention relates to the technical field of security electronic system data backup control, in particular to a method for managing and switching a high availability multi-machine backup routing table. The method is characterized in that a cluster computer system is composed of a plurality of transaction hosts, and the transaction hosts receive and process orders through a middleground access gateway and return processing results to a foreground access node through the middleground access gateway. The hosts communicate through networks, a file system is read, written and accessed through a shared storage device, consistency data among the hosts is coordinated and synchronization among the hosts is conducted through distribution lock service, and a node failure and node adding notice is obtained through cluster member management service. The method is applicable to a cluster system, a cold backup device is added, an order routing table capable of defining host logic reciprocal relation flexibly is provided, high availability, concurrent processing of various items, handling capacity, delay response, a switching time index of the system remain a level close to an original one-host one-backup-device mode, and cost is low.

The invention provides a distributed traffic signal phase difference control structure and a phase difference adjustment method. Two functions of phase difference optimal computation and phase difference adjustment are separated, the phase difference optimal computation is completed by a central computer, the phase difference adjustment relevant to the state of each road junction is completed in a local road junction traffic signal controller, a phase difference adjuster is internally installed in each traffic signal controller, a phase difference generator of the central computer figures out target phase difference of each road junction, then the phase difference generator sends each target phase difference to the traffic signal controller of each road junction, and the traffic signal controllers of road junctions concurrently perform phase difference adjustment computation according to the current phase differences. The distributed traffic signal phase difference control structure and the phase difference adjustment method have the advantages that the central computer can be concentrated on acquisition of optimal phase difference, the phase difference generating algorithm can be carefully selected and optimized, the computing power of numerous road junction controllers can be fully utilized, and simultaneously the autonomous decision-making and autonomous control capabilities of road junctions are enhanced.

The invention, which belongs to the technical fields of the intelligent networking vehicle and vehicle-road coordination, provides a cloud-service-based green-wave vehicle speed guidance method and system of light-controlled intersection. The system is composed of a cloud server, a signal light data acquisition device, and an intelligent networking vehicle client, wherein the signal light data acquisition device and the intelligent networking vehicle client are connected with the cloud server respectively. In addition, according to the method, signal light configuration and real-time state data are collected; the collected data are uploaded to the cloud server in real time and the corresponding data are stored in a database; on the basis of signal light data stored in the database, the cloud server generates signal light state data and green-wave vehicle speed guidance service data based on a client request and sends the data to the client; and the client receives the data and then displays the data for a user by an intelligent mobile terminal. Therefore, the system structure of the green-wave vehicle speed guidance application of the light-controlled intersection is simplified; the communication requirement and the computing capability requirement of the system equipment are reduced; air interface communication resources are saved; and the application deployment and promotion costs are reduced.

The invention relates to a parallel task execution method and device based on the Hive. The method includes the steps that a Hive script is operated; sequence marks of process control labels are recognized; all the sequence marks are compared, so that an execution sequence of code segments is obtained; the code segments are executed according to the execution sequence, wherein parallel execution is performed on the code segments with the same sequence mark. According to the parallel task execution method and device based on the Hive, parallel and serial relations between the code segments in the Hive scrip can be freely controlled by a developer, the execution efficiency of the Hive script is greatly improved, and the calculation capability of a Hadoop cluster can be utilized more fully.

The invention relates to air quality detection technology, and provides an air quality detection method and device. The air quality detection method and device aim at achieving wide-range detection on air quality. The air quality detection method based on an intelligent air conditioner includes the following steps that an air quality detection module of the intelligent air conditioner detects outdoor air quality parameters and transmits the outdoor air quality parameters and ID of the intelligent air conditioner to a cloud platform through a communication module; the ID sent by the intelligent air conditioner is matched through the cloud platform so that geographical location information of the intelligent air conditioner can be acquired, and conducts comprehensive calculation and analysis on the outdoor air quality parameters sent by the intelligent air conditioner so that regional air quality situations can be acquired. The invention further provides the intelligent air conditioner with the air quality detection function and an air quality detection system containing the intelligent air conditioner. The air quality detection method, device and system are suitable for air quality detection.

The present invention discloses a video analysis method and device. The method comprises: performing slicing process of video files, and obtaining video slices; distributing the video slices to a plurality of processors, performing decoding and analysis of distributed video slices by a plurality of processors through an analysis model; and integrating the analysis results of all the video slices. The video analysis method and device fully employ the advantage of parallel distributed calculation of processor and fully employ the calculation capability of each processor, so that the analysis speed is fast and the efficiency is high.

The invention discloses an offline face recognition system and an offline face recognition method based on low-power-consumption embedded and infrared triggering. The offline face recognition method comprises the steps of continuously detecting an object distance, and returning the object distance by means of a voltage value manner; determining whether an object enters a recognition range according to the returned voltage value, acquiring an image of the image which enters the recognition range, determining whether the acquired image is a face image, if not, displaying error information, and if yes, determining whether the current face image belongs to a moving object, if not, displaying the error information; and if yes, performing matching on the face image and stored information in a face characteristic database, and outputting a matching result. The offline face recognition system and the offline face recognition method have advantages of ultralow power consumption, high mute stability, small size, high convenience in mounting, offline recognition, normal operation without networking, high stability and high reliability in operation.

The invention discloses a virtual machine and a cross-OS (operating system) data processing system and method. The cross-OS data processing method includes that the virtual machine loaded and operated in the mobile terminal receives a user request processed by a desktop terminal upon mobile terminal data; the virtual machine executes processing procedures corresponding to the user request upon the mobile terminal data; the virtual machine returns processing results to the desktop terminal. The virtual machine containing a data processing system is loaded and operated on the mobile terminal, control rights of the mobile terminal and data therein are obtained through the virtual machine, when a user needs to restore the mobile terminal data or obtain evidence, the desktop terminal is responsible for sending a control request, and the virtual machine executes the data processing procedures corresponding to the request so as to achieve the purposes of restoration or obtaining the evidence. Copy is omitted and a data processing platform is transferred from the desktop terminal to the mobile terminal loaded with the virtual machine, so that execution efficiency is improved, time is saved, mobile terminal computing power is fully utilized, and execution stress of the desktop terminal is lowered.

The invention discloses a satellite mobile communication network multi-machine collaborative simulation system, belonging to the technical field of satellite network simulation. The system comprises an interaction control assembly and a user terminal assembly, a mobile satellite assembly, a network assembly and a result display assembly which are in data connection with the interaction control assembly, wherein the interaction control assembly realizes the coordination and the control among the assemblies of the system; the user terminal assembly realizes the acquisition and the transmission control of service information; the mobile satellite assembly realizes the configuration and the simulation of mobile satellite entity functions and controls the execution of a mobile satellite simulation process; the network assembly realizes the configuration and the simulation of mobile satellite network entity functions and controls the execution of a mobile satellite network entity simulation process; and the result display assembly realizes the display of predefined simulation results and dynamic simulation data. The invention additionally discloses a multi-machine collaborative simulation method corresponding to the system. The invention can be used in the satellite mobile communication field.

The present invention relates to an embedded parallel computation system and embedded parallel computation method. It includes main machine and application subsystem, said main machine has main machine processor, main machine memory and main machine bus, and said application subsystem includes application processor, application memory and application bus. In the main machine memory a main machine program including initialization routine and task interrupt routine is stored. Said application subsystem includes task control register, task condition register and multi-task operation system with task supervisor.

The embodiment of the invention provides a video splitting method and equipment and a video analysis system. A video can be split according to a split number in a video split table; and meanwhile, by the video splitting method and equipment provided by the invention, the split number in the video split table also can be modified according to a ratio of an accumulative analysis duration to a datum analysis duration. According to the invention, when the accumulative analysis duration is relatively long, the split number can be correspondingly reduced, so that a computed amount in the process of analyzing each sub video is improved, and computing power of a computing node can be sufficiently utilized; and when the accumulative analysis duration is relatively short, the split number can be correspondingly increased, so that the computed amount in the process of analyzing each sub video is reduced, and overstock of the split sub videos in the computing node is prevented. Therefore, according to the video splitting method and equipment and the video analysis system which are provided by the invention, reasonable splitting on a video is implemented, and not only computing power of the computing node is sufficiently utilized, but also overstock of the split sub videos in the computing node can be prevented.

The invention discloses a quick filtration system facing large scale video analysis. The quick filtration system facing large scale video analysis includes a multistage filter, a full characteristic network model, and a global feedback queue module, wherein the multistage filter is used for filtering background frames, frames without target objects, and frames that the target object number is lessthan the value threshold Number of Object, in video streams, so as to obtain the detected frames in the video streams; the full characteristic network model is used for detecting and analyzing the detected frames; the global feedback queue module is used for coordinating the processing rate of each filter in the multistage filter; and a queue is added between the front and back two continuous filters of the multistage filter, so as to enable the multistage filter to perform asynchronous parallel execution in a mode of a streamline. Through cascade connection of the plurality of filters, the quick filtration system facing large scale video analysis can greatly reduce the number of the video frames requiring analysis of the full characteristic model, and through the asynchronous parallel cascade structure, can improve the number of the video streams which can be detected by the system in real time in the same hardware environment.

The invention provides a distributed graph data processing method and device. The distributed graph data processing method includes: determining if a second calculation message sent from a second slave calculation node is received or not in a process of waiting to receive a reply response sent by a target slave calculation node when a first slave calculation node executes the current iteration treatment completely, it is determined that a second calculation message to be treated does not exist in a message queue, and it is determined that an active vertex does not exist in a block which is currently treated; and executing next iteration treatment if so, wherein the target slave calculation node can receive a first calculation message sent from the first slave calculation node. The distributed graph data processing method and device can fully utilize the calculation capability of slave calculation nodes, can shorten the synchronization waiting time, can optimize the execution efficiency, and can shorten the total execution time.

The invention discloses a video compression method and device. The method comprises the following steps: carrying out slicing processing on a video file to obtain video slices; distributing the video slices to a plurality of processors, and compressing the distributed video slices by the plurality of processors; and recombining the compressed video slices. By adopting the video compression method and device, the advantages of parallel distributed calculation of the plurality of processors are fully used, the computing power of each processor is fully used, the compression speed is fast, and the efficiency is high.

The invention discloses a test method and device. The method comprises the following steps of: obtaining a to-be-executed test case set; determining a distribution granularity of distributing different test cases for each queue in each period according to test cases and features of each queue in a queue set for executing the test cases; when a predetermined period arrives, detecting idle queues, not executing the test cases, in the queue set; distributing different test cases for the idle queues according to the distribution granularity of each idle queue until test cases in the correspondingdistribution granularity are distributed to each idle queue, or until the test cases in the test case set are completely distributed; and obtaining execution results of the distributed test cases fromthe queues to which the test cases are distributed. By implementing the test method and device, the test case execution efficiency can be enhanced.

The invention relates to an abnormal detection method of a machine tool spindle. The abnormal detection method of the machine tool spindle comprises the following steps of machine tool spindle variable speed no-load test, data preprocessing and feature extraction, and model training and model prediction. Variable speed no-load test data are used for performing data analysis and spindle health condition modeling; data preprocessing and feature extraction algorithms are deployed to an edge device and carried on; and principal component analysis (PCA) algorithm is used for establishing a baselinemodel of the spindle health condition, and a residual error (SPE) of a model predicted value and HotellingT2 statistical magnitude (T2)are calculated to detect spindle anomaly. The abnormal detectionmethod of the machine tool spindle can realize real-time monitoring of the machine tool spindle health condition, accurately predict spindle anomaly, and has good operability and universality.

The invention discloses a method for realizing communication between a three-dimensional image server and a client, which comprises the following steps: A1 order channel connection and data channel connection are respectively established between the three-dimensional image server and the client via a private communication protocol; A2 the client registers an order channel and a data channel to the three-dimensional image server; A3 the three-dimensional image server transmits an interactive order and operation parameters via the order channel with the client; A4 the three-dimensional image server carries out image three-dimensional reconstruction according to the operation parameters and generates three-dimensional image data; A5 the three-dimensional image server sends the three-dimensional image data to the client via the data channel; and A6 the order channel connection and the data channel connection are cut off by the three-dimensional image server and the client.

The invention discloses a method and a device for multi-channel video condensation. The method comprises the steps that a host process acquires input to-be-condensed video file identification information, and dynamically adjusts a starting way of a condensed subprocess according to the central processing unit (CPU) performances of a current computer system; the host process sends a video condensation command with the to-be-condensed video file identification information to the started condensed subprocess; and the condensed subprocess opens a corresponding to-be-condensed video file according to the command, and selects a corresponding video condensation algorithm according to the type of the to-be-condensed video file so as to carry out video condensation. Due to adoption of a multiprocess method, the number of condensed video files in unit time is increased. Through synchronous processing by the multiprocess, the CPU computing capacity of the computer system is effectively used and the internal storage use efficiency of the computer system is also improved, so that video condensation efficiency and performance use ratio of the computer system are improved.

The embodiment of the invention provides an object processing method, a distributive file system and a client device. A virtual machine can be installed in each storage server, an HTTP (Hyper Text Transfer Protocol) request containing processing parameter information transmitted by a client is received through an API (Application Program Interface) server, an address of one storage server stored with a to-be-processed object is acquired from a metadata server, and identification information and processing parameter information of the to-be-processed object are transmitted to the storage server located at the address. Since each storage server is provided with one virtual machine, the object on the corresponding storage server can be directly processed by the virtual machine which is installed on the storage server storing the object according to the processing parameter information, the downloading of the object is not needed, and more convenience and rapidness can be realized. By adopting the object processing method, the object can be rapidly processed. Meanwhile, the method also adequately utilizes the computing capacity of the storage servers, so that the utilization rate of the storage servers is increased.

The invention discloses a lightweight secrete discovery and dissemination method for wireless body area network safety data. The method includes the following stages that at the first stage, a base station performs grouping on sensor nodes, a one-way secrete key Hash chain is created for each sensor node group, and orientation Hash values of the Hash chains are distributed to the sensor nodes of the corresponding groups to serve as secrete keys; at the second stage, the base station executes preprocessing on data packets, and namely the data packets are established and then disseminated; at the third stage, after receiving the data packets, the sensor nodes recover group information included in the data packets, judge whether the received data packets are legal or not and finally process the received data packets according to a judging result. The method has the advantages of resisting sensor node compromise, being high in robustness in the aspects of packet loss and out-of-sequence of data packet transmission, not needing time synchronization between the base station and the sensor nodes and the process that digital signatures are used for guiding identity authentication and allowing dynamic data.

Embodiments of the invention provide a calculation task distribution method and apparatus, which relate to the technical field of computers and are used for improving the processing efficiency of calculation tasks of a Hadoop cluster. The method comprises the steps of obtaining a calculation resource of a calculation node, wherein the calculation resource comprises a kernel number of a central processor unit (CPU) of the calculation node and the size of an idle memory of the calculation node; according to the kernel number of the CPU of the calculation node, determining the number m of the calculation tasks of the calculation node, wherein m is the number of the calculation tasks of the calculation node; according to the number m of the calculation tasks of the calculation node, determining a memory required for the m calculation tasks of the calculation node, and determining whether the idle memory of the calculation node meets the requirement of the memory required for the m calculation tasks of the calculation node or not; and when the idle memory of the calculation node meets the requirement of the memory required for the m calculation tasks of the calculation node, allocating the m calculation tasks to the calculation node.

The invention discloses a network troubleshooting system based on a software defined network technology and a working method thereof. The system mainly comprises a message interceptor, a monitoring platform, an interactive client, a fault detection application and a probe generator. The interceptor intercepts an OpenFlow message of the controller; and after mirror image and other actions are addedto the new flow table, the data packet matched with the table item is mirrored to a monitoring platform to carry out processing processes including journey assembly, hook triggering, data statisticalstorage and the like, and a user inquires and interprets a processing result through a client and a fault detection application, so that the network monitoring and fault analysis functions are completed. According to the invention, under the existing network and hardware conditions; a solution for seamless automatic troubleshooting including a problem abstraction component, a query language component, a monitoring platform component, a fault detection application component and the like is provided, the method is suitable for a data center network environment, various common network faults canbe covered, and an API interface with good design is further provided to facilitate function expansion.

A picture processing method in an open platform includes the following steps: reading picture data; transmitting the picture data to a desktop client through a web page of the open platform; and processing the picture data in the desktop client, and passing back the processed picture data. In the picture processing method and a picture processing system in the open platform, the picture data required to be processed are transmitted to the desktop client through the web page of the open platform, processing of the picture data is achieved with the help of the desktop client, and the picture data obtained through the processing is passed back to the web page of the open platform. Due to the fact that the processing of the picture data is achieved in the desktop client, local computing power of a computer is used fully, and the picture processing method and the picture processing system in the open platform do not depend on Flash plug-ins and an open platform server and improve stability of picture processing of the open platform.

The invention discloses a singular value decomposition operation implementation method and device and related equipment, and the method comprises the steps: constructing a singular value decomposition operator which is used for carrying data in target equipment and carrying out singular value decomposition operation; deploying the singular value decomposition operator in the target equipment, and the target equipment is an Ascend AI processor; and obtaining to-be-processed data, and performing singular value decomposition operation on the to-be-processed data based on the deployed singular value decomposition operator. Compared with the prior art, in the scheme of the invention, the singular value decomposition operator capable of carrying the data in the Ascend AI processor and carrying out singular value decomposition operation is constructed, and the singular value decomposition operator is deployed in the Ascend AI processor, so that the calculation capability of the Ascend processor is fully utilized; and SVD operation is performed on the to-be-processed data directly based on the Ascend AI processor.