156results about How to "Valid data support" patented technology

The invention relates to the technical field of artificial intelligence. The embodiment of the invention provides a classroom student behavior analysis method and device, and the method comprises thesteps: collecting classroom images in a classroom in a preset time interval, and extracting a human body contour image; determining a human body contour image corresponding to each student; identifying human body key points in the human body contour image corresponding to each student in the classroom image; and determining the student behavior type of each student based on a classification neuralnetwork, wherein the classification neural network takes the human body key points and the corresponding student behavior types as input for training. Therefore, the image human body recognition technology and the neural network technology are combined, the student behavior types of students in the classroom can be quickly and accurately recognized, and effective data support can be provided forsubsequent education analysis or concentration analysis and the like.

The invention relates to an electricity stealing preventing exception analysis system. The system is provided with a data acquisition module, a data processing module, a detection prewarning module and a state diagnosis module and mainly applied to monitoring an electricity unitization operation status of industrial and commercial big customers and providing active business tool and approach for electricity stealing preventing business marketing. The electricity stealing preventing exception analysis system intends to solve the situation that the present electricity stealing preventing work cannot be effectively developed and the technical approach is difficult to implement, monitor the user current in real time, monitor the on and off states of a measuring cabinet and analyze voltage loop data of a secondary side; and the electricity stealing preventing exception analysis system is organically combined with an electric energy acquisition system to provide a comprehensive solution for power supply enterprise to check electricity utilization, fight against electricity stealing, examine leakage and press for payment afterwards; and therefore, real-time alarm is given on electricity stealing and current loop electricity stealing can be orderly followed.

The invention provides a floating car data-based urban road traffic accident influence prediction method, which comprises step 101, querying floating car historical data of a related road section, and preprocessing the floating car historical data to obtain the average car speed and car density of the corresponding road section under a normal state; step 102, querying floating car real-time data of the related road section, and preprocessing the floating car real-time data to obtain the current average car speed of the corresponding road section; step 103, querying road network data of an urban road traffic accident road section, and processing to obtain the topological structure of a surrounding road and the car density of the corresponding road section under a full-load condition; step 104, performing prediction analysis on the influence of an accident and road sections around the accident according to the preprocessed floating car historical data, floating car real-time data and road network topological relation data. According to the prediction method, the floating car data are input as predicting data, so that the accuracy of the prediction result of the method is ensured from the predicting data basis.

The invention discloses an analytical method of small-size impurities in steel. The technical scheme adopted by the method comprises the steps of: (1) electrolyzing a sample in neutral organic electrolyte under constant current and voltage; (2) filtering a solution rich in a large amount of impurity particles after electrolysis by a filter membrane with superfine aperture so as to settle the small-size impurity particles on the membrane; (3) taking off the filter membrane, bonding the membrane on a supporting table and putting the filter membrane on a scanning electron microscope for observation, taking morphology photos and analyzing energy spectrum components; and (4) processing data, and conducting three-dimensional statistic analysis to all impurities in the sample or statistic analysis to specified type of impurities. The analytical method of small-size impurities in steel solves the problems that the electrolysis extraction method of bulk sample is not applicable to the traditional analysis method of small-size impurities, the impurities cannot be displayed completely in the analysis of impurities on a two-dimensional polishing surface, and the analysis of components is interfered with a matrix.

The invention discloses equipment for monitoring the state of a current collector of a medium and low-speed magnetic suspension train. The equipment comprises a data acquisition system, an image acquisition system and a data storing and processing system, wherein the data acquisition system and the image acquisition system transmit detected motion state and current collecting state information and a motion image to the data storing and processing system; and the data storing and processing system stores and comprehensively processes the motion state and current collecting state information of the current collector and the motion image so as to output state data of the current collector. The data storing and processing system in the monitoring equipment acquires the motion state and current collecting state information of the current collector and the motion image at the same time, so that the acquired data and the acquired image can be comprehensively processed, the image can verify the detected data, and factors of the current collector, such as the material, the structural design, the mounting specification and the like, can be optimized, namely efficient data support for the improvement on the current collector is provided.

The invention relates to a big data analyzing system, in particular to a large data advertisement and big data commodity pushing and analyzing system. The big data analyzing system comprises a data collecting module, a distributed type computing architecture and a cloud computing. The data collecting module collects information such as past historical browse and purchasing records of a user and stores the information in the distributed type computing architecture; when the user opens an internet media type website or an electronic commerce type website, distributed type mining is performed on collected data through the cloud computing; advertisement information and commodities that the user is interested in are fed back.

The invention provides a database security auditing method and device and electronic equipment. The method comprises the steps of obtaining DPI database log data in real time, and obtaining an SQL statement access event according to the log data; matching the SQL statement access event with a preset user behavior rule base, wherein the preset user behavior rule base is a rule base of user behaviors obtained by performing association analysis according to historical log data and / or real-time log data of the DPI database; and if the SQL statement access event is not matched with a preset user behavior rule base or partially not matched with the preset user behavior rule base, determining the danger level of the SQL statement access event. According to the method, the association rule can beadjusted according to the structural change of the database and the new behavior of the user, the user behavior is dynamically discovered, and the auditing strategy is adjusted, so that the method adapts to different application scenarios and novel user behaviors, and an effective database security guarantee can be provided.

The invention belongs to the technical field of airplane component strength test, and particularly relates to an airplane component strength test high-temperature thermal environment simulation device, which comprises: a high-temperature thermal environment box, an inert gas inlet communicated with the internal space of the high-temperature thermal environment box and an inert gas outlet communicated with the internal space of the high-temperature thermal environment box are arranged on the high-temperature thermal environment box; an airflow diffusion nozzle which is arranged in the inert gas inlet; a micro-positive pressure relief valve which is arranged at the inert gas outlet; and a graphite heating element which is arranged in the internal space of the high-temperature thermal environment box so as to carry out radiation heating on the aircraft parts located in the internal space of the high-temperature thermal environment box. In addition, the invention relates to an aircraft component strength test high-temperature thermal environment simulation method which is implemented based on the aircraft component strength test high-temperature thermal environment simulation device.

The invention discloses a monitoring system and a method of the Internet of Things water for aquaculture. The monitoring system comprises an oxygenation device, water pumps, a collecting system used for collecting a water parameter, a wireless monitoring terminal and a monitoring server. The wireless monitoring terminal is connected with the collecting system and the monitoring server, and the wireless monitoring terminal controls the water pumps and the oxygenation device. The collecting system is one set. The wireless monitoring terminal controls the water pumps to respectively pump water of aquaculture points to a common water outlet pipe orderly, the collecting system is arranged at the tail end of the common water outlet pipe, and the tail end of the common water outlet pipe is placed in a precision improving device. The monitoring system and the method of the Internet of Things water for the aquaculture achieve prolonging the service life of the oxygenation device, and meanwhile lowering the cost of the device and power consumption. One set of wireless monitoring terminal and one collecting system are shared by multiple aquaculture points, and the comparability of the water parameters of the multiple aquaculture points is improved, and the cost of the water quality monitoring device of the aquaculture points is lowered. A sensor is arranged at the position of the common water outlet pipe, and the sensor is convenient to clean and replace.

The invention discloses an icing reservoir water temperature and ice condition inversion and prediction method based on a thermal coupling model, and the method comprises the steps: carrying out the processing of the thermal coupling model constructed by a hydrodynamic module, a water temperature module and an ice module, and considering the hydrodynamic water temperature and ice coupling effect;according to the method, the thermal evolution process of the water temperature and the ice condition of different types of frozen reservoirs in high and cold regions can be realized, so that the inversion and prediction of the distribution condition of the water temperature and the ice condition of the frozen reservoirs are achieved, and effective data support is provided for reservoir construction.

The invention discloses a rail transfer station passenger flow jam risk assessment method based on AFC (automatic fare collection) card swiping data. The method includes the steps: preprocessing AFC original transaction data; assessing rail transfer station passenger flow jam based on the AFC card swiping data; extracting, removing and screening AFC original card swiping data and recognizing transfer passenger flow based on database analysis and data mining technology; acquiring arrival, departure and transfer passenger flow of a transfer station at different time intervals based on the AFC card swiping data, and manually measuring physical parameters at typical positions in the transfer stations on site; building passenger flow saturation assessment indexes at the typical positions to comprehensively assess passenger flow demand and supply relations at the positions; calibrating model parameters according to index maximum values serving as input values of a model in assessment time intervals; calculating the weight of saturation indexes of different typical positions in the model according to an entropy weight method, clustering risk assessment and dividing rail transfer station passenger flow jam five-level risk grades.

The present application provides an agricultural product traceability method, which includes: storing basic information of a target agricultural product, inspection task information of each stage in aplanting plan, inspection task execution result information of each stage in the planting plan, transportation trajectory information of the target agricultural product, and processing information ofthe target agricultural product into a block chain; generating a unique graphic code for each independent product of the target agricultural product, wherein the graphic code includes the traceability address of the target agricultural product, and the consumer accesses the relevant information of the target agricultural product stored in the block chain by scanning the graphic code. The invention realizes that the whole process data of agricultural products from the seed stage to the final sale to the consumers is stored in the block chain. By utilizing the tamper-proof property of the blockchain, the manager and the consumers can only access the data in the block chain, but can not tamper with the data, thus ensuring the safety and reliability of the data.

The invention relates to a hazardous chemical transportation risk prediction system based on big data, which is characterized in that the system comprises a data expression module, a risk factor statistics and analysis module, a prediction module and a simulation experiment and system realization module. The invention makes full use of the interdisciplinary advantages of the internet of things technology, data mining technology, statistical management technology, mobile communication technology and other disciplines, performs in-depth study of hazardous chemical transport oriented data warehouse models, risk factor analysis methods and risk prediction mechanisms, and strive to make breakthrough in the study of key issues about the regional hazardous chemical transport risk analysis and prediction.

The invention relates to an online monitoring system for an overhead line system and a method thereof. The method comprises the following steps: establishing a plurality of distributed independent monitoring systems on the overhead line system along the railway, transmitting monitoring information to a nearby signal concentrator, and transmitting the information to a monitoring management center and a big data center through a base station; analyzing and processing the data by the monitoring management center, transmitting a monitoring command, and transmitting the command to the specified independent monitoring system by the base station and the signal concentrator. According to the online monitoring system disclosed by the invention, a real-time state monitoring scheme for key technical parameters of the overhead line system is provided, the operating state of the overhead line system is accurately, reliably and safely monitored in an all-weather manner, and an effective data support is provided for intelligent operation maintenance of the overhead line system; with the adoption of the data analysis processing center based on a big data platform, a brand-new overhead line system data analysis and operation mode is provided, the real-time data, historical data and business data can be unified and integrated, and an effective implementation method is provided for the state of the overhead line system.

The invention discloses a special-purposed fan assembling and detecting device and a circle run-out detection method. The device comprises a frame, a workbench is fixed in the top of the frame, a detection platform is fixedly mounted on the workbench via a clamping cylinder and provided with a tool clamp, the bottom of the workbench is provided with a fan driving motor, the fan driving motor is arranged below the detection platform, a laser range finding device is arranged above the detection platform, fixedly mounted on the workbench, and connected with a main controller via a data collection card, the side surface of the workbench is provided with an automatic screw installing device, and the automatic screw installing device is connected with a screw feeding cylinder. Circle run-out of a fan of an automobile air conditioner is detected to reduce noise generated in the using process of the fan, the using comfort level is improved, the service life of the fan is prolonged and the reliability is enhanced.

The invention discloses an Internet of Things based welding work management method, and relates to the technical field of welding work management. The method specifically comprises the following step:S1, technologically scanning codes through an intelligent acquiring terminal before the welding work so as to obtain basic welding work requirements, mainly including the welding technology parameterrequirements such as current, voltage, preheating temperature, interlayer temperature and welding speed, the welding qualification requirement, the welding material size requirement, and the weldingenvironment requirements such as temperature, humidity and air speed. With the adoption of the method, the welding data can be comprehensively monitored in the whole process in real time, so that datasupport is supported to management and decision making, and interference of human factors can be effectively decreased; the welding data are true and reliable; the welding quality is improved; the construction environment safety factors are monitored and alarmed in real time, so that the safety accident is reduced; the energy consumption equipment are adjusted, controlled and managed, so that theequipment utilization rate is increased; the equipment energy consumption is decreased; and welding materials are effectively managed and controlled.

Embodiments of the invention provide a query statement processing method and device, and relates to the field of databases. The method comprises the following steps of: analyzing a query statement, anexecution consumed time of which does not exceed a set threshold value, so as to obtain an analysis result; judging whether the analysis result satisfies a risk judgement condition or not on the basis of the analysis result; and if the analysis result satisfies the risk judgement condition, determining the query statement as a risky query statement. The query statement processing method and device are beneficial for providing effective and comprehensive data support for database optimization.

The invention relates to an indoor air quality evaluation system and method. The evaluation system comprises an acquisition module, a judgment module, a calculation module and a signal transmission module, wherein the acquisition module is used for acquiring a plurality of measured values of air quality parameters of a living room in current season and sending the measured values to the judgment module; the judgment module is used for judging whether the measured values are in separate-corresponding current-season preset standard ranges or not, obtaining a plurality of judgment results and sending the judgment results to the calculation module; the calculation module is used for carrying out calculation according to the judgment results to obtain an air quality comprehensive evaluation value and controlling the signal transmission module to send a corresponding preset air quality grade to a terminal device for displaying through the air quality comprehensive evaluation value. According to the system and the method, through carrying out acquisition and judgment on the air quality parameters of the current season and obtaining an evaluation value to reflect air quality of the living room through a judgment result, an effective data support is provided for comprehensive control on the air quality of the living room.

The invention relates to a hazardous chemicals transportation risk prediction method based on big data. The method comprises the following steps: aiming at the low-probability high-risk characteristics of a hazardous chemicals transportation risk event, based on risk classification and factor correlation analysis, adding weight for each risk factor, while taking into account the spatial characteristic, transportation accident rate, exposed population, and expected loss measurement factor, integrating multiple single and quantitative prediction models through an integrated approach, to construct a hazardous chemicals transportation risk multi-dimensional factor combination forecasting model, so as to effectively improve the prediction accuracy.

The invention relates to a CNNGRU water quality prediction method fusing an attention mechanism. The method comprises the following steps of cleaning and normalizing pollutant concentration, river water flow rate and water flow rate data, constructing the data into a vector form, inputting the vector form into a CNN model for feature extraction, and taking feature vectors extracted by the CNN model as input of a GRU network for model training; meanwhile, an Attention mechanism being fused in the model training process to optimize model parameters, and finally, the trained model being used forshort-term prediction of river water quality. According to the method, the strong feature extraction capability of the CNN model, the simple and easy convergence of the GRU network structure and the outstanding important features of the Attention mechanism are fully fused, and accuracy of the model is improved; effectiveness of the method is verified through specific experiments.

The invention discloses an analysis method for the overall electricity utilization trend of an electric power market, and belongs to the field of electricity supply and utilization management. The analysis method comprises the following steps: collecting and summarizing various types of electricity selling data in a district within a time period according to classes; carrying out data cleaning on the various types of electricity selling data to obtain various types of complete electricity selling data in the district; analysing the changes of the various types of electricity selling data to obtain the overall electricity utilization trend of the electric power market of the district; obtaining the development condition of the electric power market of the district and a future marketing condition predication according to the historically finished electricity selling amount historic trend data, the actual business expansion condition in the past two years and the influence amount of the extreme air temperature calculated on the basis of a historic data analysis on the electricity selling amount; predicating the potential market of electricity supply and utilization in the district by virtue of an analysis on the overall electricity utilization trend of the electric power market of the district, providing basic data for power grid optimization and marketing personnel allocation, and providing an effective data support for electric power market analysis and power grid planning. The analysis method can be widely used for the fields of electricity supply and utilization management, power grid optimization and electric power marketing.

The invention discloses a disease latent tree detection method and system based on a hyperspectral unmanned aerial vehicle, and the method comprises the steps: taking a forest region infected with a disease as a sample region, and randomly selecting a plurality of sample plants; periodically collecting leaves, acquiring spectral data of the leaves, judging the health state of each sample plant, determining a disease stress area in combination with the satellite remote sensing image of the sample area, and identifying the disease type of the sample plant; establishing a disease data model; determining a target area, and collecting a hyperspectral image of the target area; preprocessing the hyperspectral image and inputting the hyperspectral image into a disease data model to obtain a diseased plant detection result. A disease data model is established based on sample data acquired in the earlier stage, an unmanned aerial vehicle is used for carrying a hyperspectral camera for field data acquisition, the disease data model is used for detecting diseased plants, and data support is provided for large-scale forest disease and insect pest occurrence trend, disease monitoring and early warning through the hyperspectral remote sensing technology.

The invention relates to the field of remote sensing image processing, and discloses a satellite remote sensing precipitation refined space estimation method and system, so as to meet the requirementsof improving the spatial resolution and maintaining the data accuracy at the same time. The method comprises the steps of analyzing correlation coefficients between satellite annual precipitation data and all influence factors, and screening out an explanatory variable set through a stepwise linear regression method; taking satellite precipitation as a dependent variable, taking the screened explanatory variable set as an explanatory variable, carrying out regression analysis by adopting a hybrid geographically weighted regression model fusing least square method global regression and geographically weighted regression, and obtaining high-resolution regression trend surface spatial data according to a calculation result of a regression coefficient; carrying out interpolation on the regression residual error by adopting a common Kriging interpolation method to obtain residual error interpolation space data; and adding the regression trend surface and the residual interpolation spatialdata, and performing downscaling conversion according to the target spatial resolution to obtain estimated satellite annual precipitation data.

The invention discloses a bus fire accident risk assessment method. The specific assessment method comprises the three steps of 1, determining a value of a bus fire accident occurrence probability; 2,determining a bus fire accident risk level; and 3, performing bus fire accident risk early warning. The step 1 comprises the sub-steps of establishing a bus fire accident risk factor library; according to a bus running state, obtaining values of risk assessment indexes; and determining the bus fire accident occurrence probability. The six risk assessment indexes are a vehicle safety technology status, whether passengers carry flammable and combustible materials or not, a vehicle inside temperature, whether driver operation meets a driver safety operation specification or not, a vehicle insidecrowding degree and a vehicle outside environment temperature. By applying the assessment method, the bus fire accident occurrence probability can be quantitatively determined; the assessment processis simple; the operability is high; and a management basis can be provided for traffic safety management, social safety maintenance and the like.

The invention relates to an image feature point tracking method and a device. The method comprises the following steps: determining a polar line projected by a target feature point in a second image;acquiring a first gray value of a start tracking point located on the polar line; acquiring a second gray value and a gray gradient value of the target feature point; in accordance with a first gray value, the second gray value and the gray gradient value tracking the position of the target feature point in the second image in the direction of the polar line, combined with the gray value of targetfeature points, the gray level gradient value and the gray level value of the initial tracking point tracking the target feature points in the polar direction, which can increase the number of feature points successfully tracked, improve the robustness and stability of tracking target feature points, and reduce the tracking range of feature points to the polar direction, which simplifies the feature point tracking model, shortens the tracking time, improves the tracking efficiency and provides more effective data support for the subsequent image information processing such as machine vision.

The invention provides a high-speed mass formulated storage and feature preserving method for multisource unformatted broadband data. Aiming at the demand for universal data storage equipment in the development, production, test and maintenance process of weapons like unmanned aerial vehicles and stratospheric airships and civilian equipment in the field of aerospace, the technical problems like multichannel high-speed serial data reliability transmission and multisource data high-speed mass reliability storage are effectively solved by adopting a high-speed multi-channel SPIO optical fiber interface reliability transmission technology, a broadband data high-speed mass storage technology, a complicated clock management technology, a multisource unformatted data formatted recording technology, a whole-process signal feature preserving technology, a recording and accessing simultaneous supporting technology, a small-size light weight technology and the like. The high-speed mass formulated storage and feature preserving method can be applied to the development, production, test and maintenance process of the equipment like aerial unmanned aerial vehicles and stratospheric airships inthe field of aerospace, high-speed mass storage of the multisource data is effectively achieved, and key data support is provided for performance analysis of relevant equipment to achieve performancetesting and estimation for the equipment.

The invention relates to a method of information extraction for ocean eddy evolution process and belongs to the technical field of ocean information. The method is mainly applied to extract detailed evolution process information of eddies automatically from an eddy tracking data set and, technically, includes: establishing a process-oriented hierarchical data organization model, abstracting and recording the evolution process information of eddies by process, event and scenario objects, tracing the eddy evolution process by depth-first graph traversal algorithm and the like, automatically extracting detailed spatial-temporal information of process, event and process interaction, explicitly expressing and recording the process information, originally implied in the eddy tracking results, into a spatial-temporal database, and providing effective technical and data support for the analysis and discovery of eddy motion feature and law.

The invention relates to a friction-wear testing device, and particularly relates to a vacuum high-temperature high-speed friction-wear testing device. The vacuum high-temperature high-speed friction-wear testing device is provided with a hydraulic power system, a bullet, a propelling system, a sensor system, a friction pair loading system, a data acquisition system and a support frame, as well as a vacuum heating system and a sample recycling system, wherein the sample recycling system is a blind pipe with one sealed end; the open end of the blind pipe is butted with a force applying ring of the friction pair loading system; the vacuum heating system is a tubular vacuum box which wraps a section from a transmission pipe to the sample recycling system, and vacuum atmosphere and a resistance wire are arranged inside the vacuum box. The data acquisition system comprises strain gauges, a computer and a transmission circuit which is used for transmitting the data in the strain gauges to the computer; the strain gauges are adhered to both the force applying ring and an annular force meter. The device disclosed by the invention can be used for simulating a friction test in the vacuum atmosphere at different temperatures, satisfies friction-wear simulation under complex working conditions such as vacuum, high temperature, high speed and the like, and can provide effective data support for a research on friction wear under complex conditions.

The invention discloses an unsupervised change detection method and system for a homologous or heterologous remote sensing image; the method comprises the steps: inputting a homologous or heterologous dual-time-phase image into a twin full-convolution feature extraction network, and obtaining feature difference images of different scales; constructing a feature fusion network which corresponds to each scale and is based on meta learning and grouping convolution, and fusing the feature difference graphs of different scales to obtain a difference graph; using an improved objective function to complete optimization of the twin full convolution feature extraction network and the feature fusion network based on meta learning and packet convolution, and using an Adam optimization method to iteratively update weight values of the twin full convolution feature extraction network and the feature fusion network; in the training process, continuously increasing the numerical value difference between the variable pixels and the invariable pixels in the difference image, and completing binarization operation on the trained difference image through a threshold value to obtain a binary change image with the same size as the original image. According to the method, a change detection result of a pair of homologous or heterologous dual-temporal remote sensing images can be efficiently and accurately obtained in an unsupervised mode.

The invention discloses a method and system for detecting and identifying dressing information of a target person. The method comprises the steps: a, carrying out dressing image detection on the target person entering a specific region, and comparing the identified dressing image information with pre-stored dressing image information; collecting face image information of the target person enteringa specific region, and comparing the collected face image information with pre-stored face image information; and b, obtaining the dressing image information of the same target person, enabling the obtained dressing image information of the same target person and the face image information to form a one-to-one corresponding binding relationship, comparing the binding relationship with a pre-stored binding relationship, if the binding relationship is consistent with the pre-stored binding relationship, returning to the step a, otherwise, judging that the target person is an illegal person, andsending reminding information of accessing the illegal person. By means of the method, the dressing information of the person can be conveniently detected and counted; the working region informationof the target person can be effectively obtained in time, and effective data support is provided for staff management, so that the working efficiency is improved.