57 results about "Comparison diagram" patented technology

A test method for wind-photovoltaic-storage hybrid grid-connected power generation includes the following steps: acquired resource data or power data are provided for a simulated power generation source controller in order to control the output power of the simulated power generation source; the statistic of a load is provided for a simulated load controller in order to control the power of the simulated load; the initial parameters of a wind-photovoltaic-storage hybrid power generation system to be tested are provided for a monitor; the data of each part of the system is acquired, and a comparison diagram is utilized to analyze the reliability of the system; an acquired actual curve is compared with a theoretical curve so as to evaluate whether the system performance meets requirement, and the controllability of the simulated power generation source controller and the simulated load controller is utilized to carry out a reliability test on the system under extreme and faulty conditions. According to the inveniton, the economy and reliability indexes of the wind-photovoltaic-storage hybrid power generation system can be conveniently researched and analyzed.

The invention provides an air pollution assessment method and device based on a CALPUFF system. The method comprises: a data acquisition step, when acquiring an instruction for reading data, reading discharge data of at least one pollution source within a preset geographic area and air quality data monitored by a monitoring station within the preset geographic area from a local database; a data analysis step, calling a CALPUFF model interface and a CALPOST model interface, calculating pollution concentration data of spatial distribution within the preset geographic area according to the discharge data and the air quality data; and a diagram display step, displaying a concentration comparison diagram of the air quality data monitored by the monitoring station within the preset geographic area and data of influence of the pollution sources on the monitoring station according to the pollution concentration data, or displaying the pollution concentration data on an ArcGIS map in an inverse distance weighted interpolation rendering mode. The device provides execution modules for the method. The method and the device facilitate user's operation and can intuitively display air pollution data.

The present invention discloses a method for detection of foreign matters based on image identification. The method comprises: collecting images of a device to be detected and storing the images of a device to be detected; generating a standard background image according to the images, performing difference comparison of the images and the standard background image, if an all-nonzero comparison diagram is obtained, isolating foreign matter images according to a preset segmentation rule, extracting position information and area information of the foreign matters at the surface of the device to be detected according to the foreign matter images, determining whether the foreign matters at corresponding positions are pseudo foreign matters generated by light and shadow interference or not according to the position information, if the foreign matters at corresponding positions are not pseudo foreign matters generated by light and shadow interference, continuously determining whether the area information of the foreign matters is in a preset range with pixel points having the least amount of foreign matters or not, and if the area information of the foreign matters is in the preset range with pixel points having the least amount of foreign matters, determining that there are foreign matters at the corresponding positions. The method and device for detection of foreign matters can detect whether there are foreign matters at the surface of the device to be detected or not so as to facilitate timely discovering foreign matters on the device to be detected in the production process and reduce the loss because the foreign matters cannot be timely discovered. The present invention further discloses a device for detection of foreign matters based on image identification corresponding to the method mentioned above.

The invention provides a new karst stage judgment method. The karst reservoir is a new field of recent oil and gas exploration and development, and the judgment of karst development stage is the difficulty in a karst reservoir modeling technology at home and abroad. The existing karst stage judgment method mainly takes an exposure marker as basis of judgment, but various karst process forms exist, so that the karst processes caused by factors such as hydrothermal solution are difficult to judge due to the lack of an exposure marker. The invention provides a new method for objectively judging the karst stage by combining a method for judging the filling stage by carbon and oxygen isotopes of karst fillings and a method for judging the karst fluid active stage by inclusion homogenization temperature aiming at the problem of karst stage judgment. A carbon and oxygen isotope discrimination chart and an inclusion homogenization temperature comparison diagram are established by the new karst stage judgment method, so that the karst development stage can be accurately and rapidly judged, the limitation that the traditional karst stage judgment takes the exposure marker as basis is broken through, and the problem that karst stage judgment is difficult under the condition that the underground deep layer is short of exposure marker can be solved.

The invention aims to provide a method for calculating mid-frequency transmission loss of subsystems and optimizing acoustic packages. According to the method, a model for calculating the transmission loss of the whole subsystems is established on the basis of Hypermesh software and VA One software, the transmission loss of the whole subsystems is obtained through calculation, the transmission loss of each subsystem and the transmission loss of the whole are compared, whether each subsystem belongs to a weak area or a non-weak area is determined, and the acoustic packages are optimized according to an optimization objective and a result whether each subsystem belongs to the weak area or the non-weak area. Compared with the prior art, the method adopts a hybrid FE-SEA (finite element-statistical energy analysis) method, a finite element model is divided into a plurality of FE subsystems, a comparison diagram of a mid-frequency transmission loss curve of each FE subsystem at the mid-frequency band and a transmission loss curve of the whole is obtained rapidly and accurately through calculation, whether each subsystem belongs to a weak area or a non-weak area is determined and the FE subsystems are subjected to different arrangement on the basis of the optimization objective, so that the purpose of effective optimization design of the acoustics packages can be achieved.

The invention discloses an FOA optimization based GRNN (General Regression Neural Network) fault forecasting method for a rotary machine and relates to a technical field of rotary machine fault forecasting methods. The method includes steps of (1), adopting Cincinnati antifriction bearing experiment data; (2) dividing discrete data of collection signals in step (1) into two parts including training data and predication data, putting the training data into the FOA optimization based GRNN for forecasting model training; (3) performing fault forecasting on forecasting data by using the forecasting model trained in step (2), calculating the root-mean-square error and forecasting time and acquiring a comparison diagram of a simulation forecasting curve and an actual curve; (4), determining fault generation time and fault types according to results obtained in step (3). According to the invention, through optimization on GRNN smoothing factors by FOA, the FOA optimization based GRNN fault forecasting model is established, the GRNN optimal model establishing time is reduced and influence on forecasting results by human factors is reduced effectively.

The invention provides a three-directional geology modeling method based on sparse borehole points, and belongs to a three-dimensional geology modeling method for coal mine geology. The modeling method comprises steps of 1: constructing borehole database, extracting responding base information, generating comparison diagram of a complete drilling layer and an incomplete layer and using the interpolation algorithm of fixed points; 2: setting mark points of transverse virtual drilling holes and forming the whole drilling hole by using the coordinates as input parameters of interpolation based on a prospecting line profile map according to exploration data of stratum trend; 3: using crosscheck as a statistical means to test precision of models according constructed models; and 4: referring to the contour map of one stratum, comparing corresponding stratum data of models and summarizing similarity until the models are modified to a proper confidence level. In the general background of sparse borehole points of a model construction process, other non-borehole data are added, missed borehole data are virtualized, and according to possible errors of constructed models, judging abruptly changed construction methods, thereby improving construction quality of the models.

The invention discloses a nondestructive detection method of a composite insulator. The nondestructive detection method of the composite insulator comprises the following steps: coating the surface of a to-be-detected composite insulator with an ultrasonic coupling agent; outputting a pulse signal by an ultrasonic testing system; uploading the pulse signal to an ultrasonic transducer by the ultrasonic testing system and converting the pulse signal into an ultrasonic signal; transmitting the ultrasonic signal by the ultrasonic transducer to the to-be-detected composite insulator to generate an ultrasonic echo signal; receiving the ultrasonic echo signal by the ultrasonic testing system, converting the ultrasonic echo signal into an electric signal and determining resonant frequency of the electric signal; and analyzing the viscidity condition of the interface of the to-be-detected composite insulator by the ultrasonic testing system according to a resonant frequency and preset resonant frequency comparison diagram. According to the embodiment of the invention, the composite insulator not containing macroscopic air gap can be subjected to nondestructive detection.

The invention discloses a digital heating pipeline monitoring system which comprises an industrial control upper computer and a plurality of detection nodes. Each detection node comprises a plurality of temperature sensors and a single-chip lower computer. The industrial control upper computer and the detection node respectively comprise a set of wireless transceiver circuit. Each temperature sensor is used for detecting the temperature of a heating pipeline, and transmitting detected temperature signals to the single-chip lower computer. The single-chip lower computer conduct pre-processing on the received temperature signals, and then the detected temperature signals are transmitted to the industrial control upper computer through the wireless transceiver circuit. The industrial control upper computer receives temperature data transmitted by each detection node in real time, the temperature data are displayed onto a pipe network diagram, the temperature valve and changing valve of each sensor are recoded at the same time, internal temperature distribution profiles of the detection nodes are generated, and temperature variation comparison diagrams of the temperature sensors are generated. The digital heating pipeline monitoring system can rapidly and effectively identify fault points.

The invention discloses a road effective earth volume calculation method based on an oblique photography technology, and the method comprises the steps: carrying out the photographing of a road regionthrough an unmanned plane, obtaining preliminary oblique photography data, and generating an initial oblique photography three-dimensional live-action model; inputting the elevation data into a roaddesign model; generating an initial roadbed cross section diagram in a road design model according to the elevation data; generating a constructed oblique photography three-dimensional live-action model after the road construction is completed; replacing the elevation data in the road design model with the constructed elevation data to generate a constructed roadbed cross section diagram; generating a roadbed cross section comparison diagram according to the same pile number, and calculating the cross section areas of effective excavation and earth filling squares; and according to the cross section areas and distances of the effective excavated and filled earthwork squares, calculating the effective excavated and filled earthwork squares, and taking the sum of the effective excavated andfilled earthwork squares of each road section as the effective earthwork volume of the road. According to the method, the scientificity and reasonability of road earth volume calculation are improved,and a reliable basis is provided for budget and settlement of road engineering.

The invention provides a through-hole design system which comprises a design interface display module, a calculating module and an output module, wherein the design interface display module is used for displaying a through-hole design interface on a calculating device; the interface comprises a parameter input area and a result display area; the parameter input area is used for inputting a to-be-observed frequency range, a design frequency point, a panel dielectric coefficient Dk, the length Lvia of the through hole excluding a through-hole remained section, the length Lstub of a through-hole remained section, radius r of the through hole, distance S of differential through holes, radius W of an avoiding hole and reference impedance Z0; the calculating module is used for calculating practical impedance Zvia and ideal impedance Zc of the through hole according to the parameter inputted to the parameter input area and a preset formula and drawing an impedance comparison diagram according to the calculated practical impedance Zvia and ideal impedance Zc; the output module is used for outputting the impedance comparison diagram to the result display area. After the through-hole design system provided by the invention is adopted, the response speed is high and a designer can quickly find the optimum design for the through hole according to the result displayed in the result display area.

The invention relates to a generation method and device of an inter-well reservoir three-dimensional comparison diagram. According to the three-dimensional contrast method of inter-well reservoirs, based on the three-dimensional coordinates of drilling trajectories and de-lamination points above wells, inter-well comparison connection lines drawn under a two-dimensional section are converted into three-dimensional vector connection lines in a three-dimensional space, and a three-dimensional grid diagram is generated; after all inter-well comparison is completed, the inter-well comparison connection lines are classified, and inter-well comparison connection lines which are connected with one another are classified as one category; each category is corresponding to an independent reservoir unit; and the boundaries of the inter-well comparison connection lines are calculated, reservoir unit interfaces are obtained through surface fitting, and the quantitative characterization of the reservoir unit interfaces can be realized. With the method adopted, the accuracy and efficiency of the comparison of the inter-well reservoirs can be improved, and the integration of reservoir comparison, three-dimensional visualization and quantitative interface characterization can be realized.

The embodiments of the application disclose a life management system, medium and method based on iris verification, and relates to the technical field of big data analysis. The method herein includes the following steps: acquiring a user's a plurality of iris information for identify verification; analyzing the plurality of iris information, obtaining an iris changing trend chart of the user; acquiring a plurality of iris changing trend comparison diagram atlases; based on the iris changing trend chart, matching one or a plurality of iris changing trend comparison diagrams; analyzing the user's health data; and generating the user's life management plan. According to the application, the life management system, the medium and the method use big data technologies, draw and analyze iris changing trend charts so as to make a life management plan and guarantees better life state.

The invention discloses a reconstruction or extension road earth volume calculation method based on an oblique photography technology, and the method comprises the steps of carrying out the photographing through an unmanned plane, generating preliminary oblique photography data, and forming an oblique photography three-dimensional live-action model; inputting the road initial elevation data into aroad design model to generate an initial roadbed cross section diagram; after the road construction is completed, building an oblique photography three-dimensional live-action model after the construction; replacing the initial elevation data of the corresponding position in the road design model with the constructed elevation data; generating a constructed roadbed cross section diagram; generating a roadbed cross section comparison diagram according to the same pile number, determining the pile spacing of the roadbed cross section comparison diagram, calculating the earthwork cross section area, calculating the road excavation and earth filling volume according to the adjacent earthwork cross section area and distance, calculating the road surface clearing earthwork volume, and countingthe road excavation volume to obtain the excavation and earth filling volume of the reconstructed and expanded road. According to the method, the earthwork volume calculation efficiency and accuracy are improved, and a reliable basis is provided for road construction earthwork volume settlement.

The invention provides a method for analyzing the influence of interlayer thermal resistance on the stress of an asphalt pavement structure, and belongs to the technical field of asphalt pavements. The problems that an existing statistical analysis method for researching the influence of interlayer thermal resistance of an asphalt pavement structure on pavement mechanics is large in limitation, finite element analysis cannot highlight the characteristics of porous granular materials of the asphalt pavement, a modeling unit is single, and aggregate contact geometrical characteristics cannot bechanged are solved. The method comprises: using discrete element software for numerical simulation of an asphalt pavement structure; secondly, setting model boundary conditions, particle parameters, aunit contact model and a contact thermal resistance coefficient, establishing an asphalt pavement structure discrete element thermal analysis model, and performing thermal analysis; and finally, applying a vehicle load to the discrete element thermal analysis model of the asphalt pavement structure, recording stress data of the bottom of each layer at the center and the periphery of the load, drawing a stress comparison diagram according to the stress data under different interlayer thermal resistance conditions, and analyzing the influence of the interlayer thermal resistance of the asphaltpavement structure on the stress of the asphalt pavement.

The invention discloses a refined monitoring system for power supply reliability, comprising an index monitoring module, an index processing module, an index comparing module and an index sensitivity analyzing and predicting module, wherein the index monitoring module monitors indexes; the index processing module processes all types of indexes; the index comparing module performs longitudinal comparison and transverse comparison on the indexes; and the index sensitivity analyzing and predicting module performs sensitivity analysis according to the longitudinal comparison diagram of the indexes. The refined monitoring system for power supply reliability can realize layered monitoring of data, real-time data acquisition and a thorough concrete evaluation mode.

The invention discloses a selection method for epoxy mud thickness during measurement of cable skin temperatures. The selection method comprises the following steps: (1) preparing multiple pieces of epoxy mud for an experiment; (2) respectively laying the multiple pieces of epoxy mud at different positions of each of cable skins; fixing a thermocouple to each epoxy mud; respectively measuring the temperatures of the cable skins on which the epoxy mud different in thickness is laid under the same current by utilizing the thermocouples during the experiment until the cable skin temperatures reach a steady state to obtain the skin temperatures of multiple sets of experiment cables in the steady state; (3) performing skin temperature measurement in the steady state on parts, on which the epoxy mud is not laid, on the experiment cables by utilizing an infrared thermometer, wherein cable skin temperature data obtained through the skin temperature measurement is used as a temperature reference value; (4) comparing the skin temperatures, which are measured by the thermocouples, of the multiple sets of cables when reaching the steady state with the temperatures measured by the infrared thermometer, drawing up a temperature comparison diagram, and determining the optimal laying thickness of the epoxy mud. The optimal laying thickness of the epoxy mud can be reasonably selected by the selection method disclosed by the invention.

The invention discloses a drop resistance intelligence promoting bottle warmer. The drop resistance intelligence promoting bottle warmer comprises a cavity, a sleeving opening, a temperature sensing color changing layer, a drop resistance layer, a scale window, a temperature displaying bar and a music playing device, and the inside of a bottle warmer body is a cavity; the temperature displaying bar, the drop resistance layer, the temperature sensing color changing layer and a heat preserving layer are arranged on the bottle warmer body from outside to inside in sequence; the scale window is arranged in the middle of the outer side of the drop resistance layer; the music playing device is arranged at the bottom of the bottle warmer body. A comparison diagram of different patterns and temperature areas is marked on the temperature displaying bar, and therefore a parent is helped to recognize the temperature of milk; the temperature sensing color changing layer is coated with multiple patterns, the patterns of the temperature sensing color changing layer can be changed with the change of the temperature of added water, the interest is added, and the intelligence of a child is improved; the music playing device can play wonderful music suitable for an infant to listen to, the digestion by the child can be promoted, the intelligence of the child is improved, and the development of the brain is promoted; the drop resistance layer can prevent a milk bottle from falling on the ground and being broken.

The invention discloses a PIM-1-based mixed matrix membrane and a preparation method thereof, and belongs to the technical field of membrane materials. According to the preparation method, an intrinsic microporous polymer PIM-1 is taken as a membrane main body, a metal-organic framework material MOF-808 (Zr) is taken as an additive, and a solution casting-solvent evaporation method is adopted to prepare the PIM-1-based mixed matrix membrane. The preparation method comprises the following steps: preparing the MOF-808 (Zr), preparing the PIM-1 and preparing the MOF-808 coated PIM-1 membrane. A permeability comparison diagram shows that MOF-808 coated PIM-1 mixed matrix membranes with different composition proportions have certain permeability selectivity to ethyl acetate / N2, and when the doping amount of MOF-808 is controlled within the optimal range, the permeability rate of ethyl acetate is increased, and the permeability selectivity is improved. The PIM-1-based mixed matrix membrane can be used for treating VOCs-containing waste gas.

The invention belongs to the technical field of oil field development fracture horizontal well parameter optimization, and discloses a method for determining a horizontal well fracture parameter based on PEBI (PErpendicular BIsection) mesh encryption. The method comprises the following steps that: designing a target block physical property parameter; on the basis of reservoir characteristics, establishing a target block geologic model; through an NWM (Near-Well Model) module in eclipse software, endowing the fracture mesh with a value; obtaining an influence comparison diagram of different fracture parameters for a recovery ratio; and determining a reasonable fracture parameter. The PEBI mesh encryption is adopted around a shaft, and the shaft can be encrypted into other PEBIs including hexagon, cylinder and the like; a hexagonal PEBI mesh is adopted, an artificial fracture of any direction can be formed, and the accuracy of a simulation result is improved; a mesh block of which the width is far smaller than the radius of the shaft replaces the fracture so as to reflect the authenticity of a formation fracture; and meanwhile, a skin factor near the shaft is set as a negative number, and a facture situation is truly simulated.

The invention discloses a method for quickly detecting available nitrogen in soil and ammonia nitrogen in water. The method includes preliminarily preparing available nitrogen quick test paper and available nitrogen concentration color gamut comparison diagrams; dissolving the soil in water and manufacturing to-be-detected soil solution; smearing a layer of sodium hydroxide solution with the pH (potential of hydrogen) of 11-13 on the available nitrogen quick test paper; smearing the to-be-detected soil solution on the available nitrogen quick test paper; allowing the available nitrogen quick test paper to stand still for 3-5 minutes; finding the most approximate color gamut from the color gamut comparison diagrams according to displayed colors so as to obtain the range of the content of the available nitrogen in the soil. The method has the advantages of speediness, simplicity, convenience, low cost and easiness in operation. Besides, farmers can learn to use the method within a few minutes without professional knowledge, expensive instruments and complicated operation, can quickly know the range of the content of the available nitrogen in the soil within a few minutes and can be guided to reasonably scientifically apply fertilizers, accordingly, fertilizer wastage and water pollution can be reduced, and environments can be protected.

The invention discloses a runoff simulation method and system based on a hydro-meteorological time sequence. The method comprise the following steps: a runoff simulation model is established based onthe principle of a GR4J hydrological model; the runoff simulation is carried out on the drainage basin by utilizing a hydro-meteorological time sequence; the runoff simulation result with the time sequence can be directly used for analysis and drawing, a hydrology worker can know the specific time corresponding to each data result without post-processing the result, and the problem of analysis inconvenience caused by the fact that a traditional hydrology simulation result does not have the time sequence is solved. According to the invention, the model parameters and the drainage basin state ofeach time period can be automatically determined only by inputting the hydro-meteorological observation time sequence and the set rate periodic time, and the observation and runoff simulation time sequence comparison diagram of the rate periodic time and the verification period can be obtained at the same time. According to the invention, the automatic treatment of runoff simulation is realized,and the runoff simulation work of one drainage basin can be efficiently completed.

The invention discloses an effective earth volume calculation method for a poor foundation road, and the method comprises the steps of collecting a road position and survey data, obtaining terrain elevation data, building a road original terrain curved surface and a road design model, and generating an initial roadbed cross section diagram; obtaining elevation data of a construction area after road construction; replacing the original terrain curved surface elevation data with the constructed elevation data; establishing a topographic curved surface to be trimmed; replacing an original terraincurved surface in the road design model, generating a repaired and measured roadbed cross section diagram, generating a roadbed cross section comparison diagram according to the same pile number, calculating an effective excavation and filling earthwork area between an original terrain ground line and a repaired and measured terrain ground line, and calculating an earthwork desilting area and a rubble landfill area according to marks of unfavorable geology; according to the cross section area and the length of the adjacent cross sections, calculating the effective soil excavation and fillingquantity and the unfavorable geology desilting and landfilling engineering quantity. According to the method, the scientificity and reasonability of road earth volume calculation are improved, and aneffective and reliable basis is provided for budget and settlement.

The invention discloses an improved distributed auction Ik-SAAP algorithm for task distribution in a WSAN, and the method comprises the following steps: 1), building an executor cooperation task distribution mathematic model; 2), initializing network parameters; 3), selecting a decision node, and initiating an auction node; 4), setting a leapfrogging number; 5), calculating a utility value of the auction node, and returning the utility value to the decision node; 6), judging whether the decision node meets an optimal executor node condition or not: returning to step 4) if the decision node does not meet the optimal executor node condition, or else, finishing the one-time task distribution; 7), carrying out a current task through a selected executor node; 8), executing the Ik-SAAP algorithm for 200 times, obtaining a data package forwarding number comparison diagram, a remaining energy variance comparison diagram and a mean task completion time comparison diagram through simulation. The method can reduce the forwarding number of data packages of the executor node under the condition of guaranteeing the high task distribution rate, shortens the task completing time, and balances the energy consumption of the WSAN.

The invention provides an automatic uncertainty analysis platform and method based on large-scale CFD parallel computing software. The automatic uncertainty analysis platform comprises a user interaction module, a database, a server, a CFD test module, an NIPC, a post-processing module and a monitoring module. The database is connected with the user interaction module and the server, the server module is connected with and controls the CFD test module, the monitoring module and the NIPC and post-processing module, the CFD test module comprises sample generation, batch processing and automatic calculation, and finally a plurality of calculation results are output to the NIPC and post-processing module. And after uncertainty analysis and result post-processing, an analysis result and original data are returned to a database and finally returned to the user interaction module. According to the method, automatic uncertainty analysis of large-scale CFD software is completed through the processes of network access, automatic example and sample loading, cluster calculation submission, real-time progress monitoring, NIPC uncertainty analysis, automatic post-processing and the like, and experimental comparison diagrams, analysis reports, summary reports and the like are provided.

The invention discloses a stereo matching method, a terminal and a storage medium, and the method comprises the steps: sequentially selecting a pixel point in a reference image as a to-be-matched point, and calculating the gray average value of all neighborhood pixel points in a fixed window with the to-be-matched point as the center; respectively calculating a first bit string of the point to be matched and a second bit string of a first candidate matching point in a preset parallax search range in the comparison graph according to the gray average value; selecting a search strategy, calculating second bit strings of the remaining candidate matching points in the comparison graph, and respectively calculating matching cost values of the candidate matching points according to the first bit string and a plurality of second bit strings; selecting the minimum matching cost value, and taking the parallax of the candidate matching point corresponding to the minimum matching cost value as the optimal parallax of the point to be matched; and obtaining a parallax image according to the parallax of all the pixel points in the reference image. According to the invention, the resource consumption of the algorithm can be effectively reduced, and the algorithm can be deployed on a low-cost computing platform.

The invention discloses an audio starting point labeling method and a device thereof, and the method comprises the steps: obtaining an audio source file, carrying out the time-frequency analysis of the audio source file, and obtaining a spectrum energy diagram; carrying out smooth filtering on the spectrum energy diagram; establishing an oscillogram by taking the time domain in the audio source file as an abscissa and the amplitude as an ordinate, aligning the time domain of the abscissa of the oscillogram and the frequency spectrum energy diagram subjected to smooth filtering, and splicing upand down after alignment to obtain a comparison diagram for display; generating starting point labeling information according to the displayed comparison graph; preliminarily determining a pluralityof corresponding starting points in the audio source file according to the plurality of received time points, and judging and deleting an error starting point in the plurality of starting points; andlabeling the audio frame corresponding to the time point in the audio source file as a starting frame according to the finally determined time point corresponding to the starting point, and exportingthe starting frame as a labeling sample. The method has the advantages of high labeling accuracy and high generalization.