578results about How to "Solve inaccurate" patented technology

The invention relates to the technical field of personalized recommendation, in particular to a commodity recommendation method and device. The method includes: acquiring the historical purchasing records of a user according to the user information of the user; determining the selling areas which the user pays attention to; displaying recommended commodity information on the display area of a selling cabinet according to the historical purchasing record and the selling areas. By the method, the commodity recommendation method based on the selling cabinet is enriched, the commodity information recommended to the user meets the real-time requirements of the user, the recommendation results are accurate, and user experience is improved.

The invention discloses a user attribute information acquiring method and device. An embodiment of the method comprises steps as follows: acquiring track data which comprise time information and position information of multiple track points; pre-processing the track data to acquire to-be-processed track data; analyzing correlation of the to-be-processed track data to extract a correlated track data set of the to-be-processed track data; calculating similarity between the to-be-processed track data and the correlated track data set; determining user attribute information corresponding to the to-be-processed track data according to the similarity. With the adoption of the method and the device, the accuracy of the user attribute information acquired on the basis of the track data is improved.

The application of the invention discloses a project recommendation method and a project recommendation system. The method comprises the steps of acquiring history data of a user, wherein the history data comprises a corresponding relation between the user and the projects; calculating an association retrieval relevance of any two projects according to the history data; as to each project, respectively determining a preset first amount of projects, which have the maximal association retrieval correlation with the project, as association retrieval related projects of the project; building an original user-project scoring matrix according to the history data of the user; zeroing the original user-project scoring matrix by the association retrieval related projects of each project to form a calculation user-project scoring matrix; and determining a project recommendation collection of the user based on the calculation user-project scoring matrix. According to the method and the system, the accuracy of the recommendation result is improved, and therefore, data transmission speed between an electronic commerce website and a user terminal is boosted further.

The invention relates to a general fixing device for a probe of an acoustic emission detection apparatus, and particularly, relates to a general fixing device for fixing the probe. The device comprises a main body, a height-adjusting push rod, a push rod pressing cap connecting member, and a replaceable probe pressing cap, wherein the replaceable probe pressing cap has different diameters, and has a probe pressing surface which can be replaced according to the diameter of the probe; and the device further comprises a pressure display unit which is used to display a pressure signal detected by a pressure-sensitive sensor for real-time understanding of the fixing condition of the probe by detection personnel. The device of the invention has a simple and compact structure, and is convenient for carrying; the device has strong generality, and can fix according to probes with different heights and diameters; the device is convenient for operation, and is convenient for probe pressing cap replacement; the spring in the invention provides a stable pressure; therefore, even if the fixing device itself becomes loose slightly, the fixing force applied to the probe dose not change greatly, and excessively small coupling force between the probe and an object to be detected is not caused finally.

The embodiment of the invention discloses a face detection model training method, a face key point detection method, a device, a device and a storage medium. Performing three-dimensional reconstruction on the training face image based on a preset three-dimensional face model to obtain a training three-dimensional face model; Generating a training UV coordinate map including three-dimensional coordinates of the training three-dimensional face model according to the training three-dimensional face model; The training face image and the training UV coordinate diagram are used for training the semantic segmentation network, so that that face detection model is obtain, As that embodiment of the invention doe not need to manually label the train face image and the training UV coordinate map, Thepresent invention solves the problem that the training data need to be manually estimated and labeled, which leads to inaccurate training data and inaccurate coordinates of the key points of the faceoutput by the CNN, and improves the performance of the face detection model and the accuracy of the key points detection of the face.

The invention relates to a multi-spectrum auto fluorescence fault imaging reconstruction method based on single view, which belongs to the optical molecular imaging field. The method is based on the diffusivity equation model, considers the uneven characteristics of small animal body, and simultaneously also considers the spectrum and the practical application characteristics of the auto fluorescence light source. In order to achieve the purpose, the multi-spectrum auto fluorescence fault imaging reconstruction method based on single view comprises the flowing steps: firstly, data acquisition; secondly, the dispersing processing of finite element; thirdly, the optimization of the selection of the feasible light source zone; fourthly, the reconstruction of the light source. The method overcomes the defects that the reconstruction light source of the auto fluorescence fault imaging reconstruction method is inaccurate, the reconstruction speed is low, the real-time processing is not convenient, and error can exist during the multi-spectrum data acquisition.

The invention discloses an e-commerce platform commodity recommendation algorithm, which comprises the following steps of: 1, obtaining operator original data: obtaining a user and a retrieval recordcopy from an operator big data system, including a user ID, a user number, keyword content, a retrieval date, retrieval times, commodity purchase information and commodity purchase times; The invention discloses an e-commerce platform commodity recommendation algorithm. The method comprises the steps of obtaining information of a user or purchase records of other users for judgment; related commodities are pushed; pushing of regional seasons and hot commodities by the user can be met; In addition, according to the invention, the structure is simple;, if the user does not express enough interest in the information pushed for the first time, the information is pushed for the second time; and according to the current season information and a plurality of pieces of commodity information in theregion where the user is located, the push content is adjusted, so that the problem of inaccurate information during first-time push is solved in time, the commodity selection time of the user is effectively shortened, and the user experience is improved.

The invention discloses a video processing method, a device and a system. The video processing method comprises acquiring expression information when a user watches a video, recording a playing time point of a video corresponding to the expression information, performing identification on the video through the recorded playing time point to obtain video identification information used to identifying a playing time point corresponding to a video clip of the video, and displaying identification information of the video. Through the invention, the problem of inaccuracy of identification of the video clip of the video in the prior art is solved, and then the effect that the video clip of the video is accurately identified is achieved.

Provided is an autonomous driving vehicle, and an infrastructure for supporting autonomous driving thereof. The autonomous driving vehicle of the present invention receives information obtained by converging sensor information (for example, infrastructure sensor information and traffic information) based on an infrastructure and vehicle sensor information selected by the infrastructure from the infrastructure positioned at the outside. Accordingly, the autonomous driving vehicle of the present invention may overcome a limitation of vehicle-based autonomous driving, thereby improving obstacle avoidance performance, solving inaccuracy of a sensor, and decreasing frequent setting of a path.

The present invention relates to a roadside parking space detection device and a detection method thereof. The roadside parking space detection device is characterized in that the device includes a geomagnetism sensor, a radar sensor, a parking space detection processor and a power source management module. The detection method includes the following steps that: the geomagnetism sensor detects geomagnetic field signals on a parking space and transmits geomagnetic field change waveforms to the parking space detection processor; the parking space detection processor judges whether the geomagnetic field change waveforms are ideal waveforms, if the geomagnetic field change waveforms are the ideal waveforms, the parking space detection processor determines a parking condition on the parking space according to a geomagnetic detection mechanism; if the geomagnetic field change waveforms are not the ideal waveforms, a radar detection mechanism is started; and the parking space detection processor determines the parking condition on the parking space according to the detection result of the radar detection mechanism. With the roadside parking space detection device and the detection method thereof of the invention, the organic combination, mutual support and mutual dependence function of the geomagnetism sensor and the radar sensor are realized, and the problem of the inaccuracy of geomagnetic parking space detection in extreme cases can be solved, and the problem of too much power consumption of the radar sensor can be also solved.

The invention discloses a statistical method and device for visitor behavior data. The method comprises the following steps: acquiring visiting behavior data of a user visiting a page; establishing an anonymous ID (Identity) for a user according to the visiting behavior data; extracting a user ID from login behavior data of the user under the condition of logging in the page by the user; establishing an incidence relation between the anonymous ID and the user ID; and obtaining the visitor behavior data of the user according to the incidence relation. With the adoption of the statistical method and device for the visitor behavior data, the problem in the prior art that the visiting data of the user cannot be accurately analyzed by using the real ID is solved and the effect of accurately obtaining the visitor behavior data is realized.

The embodiment of the invention provides a method and a device for determining a switching path between satellites, a terminal and a storage medium. The method comprises the following steps: determining residence time information corresponding to the target terminal passing through the target terminal in respective signal coverage ranges of a plurality of over-the-top satellites based on satelliteephemeris corresponding to a plurality of satellites and position information and speed information of the target terminal; generating a plurality of inter-satellite pre-switching paths for the target terminal according to the residence time information corresponding to the target terminal in the respective signal coverage ranges of a plurality of over-the-top satellites; determining inter-satellite switching times and feed switching related information corresponding to a plurality of inter-satellite pre-switching paths; and determining a recommended switching path according to the inter-satellite switching times and the feed switching related information. According to the embodiment of the invention, the method solves a problem that the finally determined recommended path is too long inswitching time delay and even is interrupted in service because only the switching times between the satellites are considered in the prior art.

The embodiment of the invention discloses a ground insulation impedance detection method, a circuit and equipment with the circuit. The circuit consists of a first voltage division circuit and a second voltage division circuit, wherein the first voltage division circuit is connected between a first output terminal and a ground wire; and the second voltage division circuit is connected between a second output terminal and the ground wire. The ground insulation impedance detection method comprises the following steps: enabling the resistance ratio of the first voltage division circuit and the second voltage division circuit to be changed from N to 1 / N to N; respectively acquiring voltages at two ends of the first voltage division circuit and the second voltage division circuit under three circuit states; obtaining a first equation, a second equation and a third equation with the detected ground insulation impedance as an independent variable according to a series connection voltage division principle; respectively solving to obtain ground insulation impedance values of the first output terminal and the second output terminal by the simultaneous equations of the first equation, the second equation, and the third equation; and taking a minimum value as the detection result. The equipment with the circuit comprises a junction box and an inverter. According to the ground insulation impedance detection method provided by the invention, the detection accuracy of ground insulation impedance is improved and the problem that the detection result of a traditional detection scheme is inaccurate is solved.

The embodiment of the present invention discloses a vehicle license plate recognition method, apparatus and device, and a computer readable storage medium. The method comprises the steps of: obtainingvehicle images collected by an image collection device, and employing an HSV color space model to generate gradient gray images of the vehicle images; in the gradient gray images, employing a projection method to determine positions of the vehicle license plates and separately extract vehicle license plate area images; according to the area of the vehicle license plate area images and the numberof the characters included in the vehicle license plates, performing segmentation extraction of the characters in the vehicle license plate area images; extracting grid features and cross point features of the characters to form recognition feature vectors of the characters; and performing feature matching in a pre-stored character library according to the recognition feature vectors of the characters to obtain the recognition result of each character, wherein the character library comprises characters meeting a vehicle license plate requirement and feature vectors corresponding to the characters. The vehicle license plate recognition method, apparatus and device, and a computer readable storage medium can rapidly and accurately locate the vehicle license plate area and can improve the recognition efficiency and the accuracy of vehicle license plate recognition.

The invention discloses a rocket and antiaircraft gun operation control and safety monitoring system and an operation monitoring method. The system comprises a plurality of operation point acquisition and control units 1 and an artificially-influenced weather operation central management platform, namely an upper computer 2, wherein data is transmitted between the upper computer 2 and the operation point acquisition and control units 1 through mobile communication and the Internet. The operation monitoring method comprises the following steps of: controlling the launching of a rocket or an antiaircraft gun; automatically acquiring operation information such as a launching pitch angle, a launching azimuth angle, rocket or antiaircraft gun launching information, longitude and latitude coordinates of operation points, operation time, operation ammunition using amount, operation channel numbers, ignition circuit resistance and the like; and uploading the operation information to a management center to monitor operation safety in a unified mode. By the system and the method, the problem of inaccurate information caused by manually recording and transmitting the operation information through personnel in the prior art is solved, and an effective tool is provided for a manager to accurately and timely master the information and scientifically perform artificially-influenced weather command and management.

The invention discloses a prediction method and device of the day runoff volume of a snowmelt period. The method comprises the steps that the variable data and the measuring parameter data of an n day snowmelt zone drainage basin in the current snowmelt period are obtained; the water-recession coefficient kn + 1 of an n + 1 day snowmelt zone drainage basin in the current snowmelt period is obtained; according to the variable data and the measuring parameter data of the n day snowmelt zone drainage basin and the water-recession coefficient kn + 1 of an n + 1 day snowmelt zone drainage basin, the water amount of snowmelt and rainfall in a drainage basin area A of the n + 1 day snowmelt zone drainage basin and the water-recession flow of river channel runoff in the drainage basin are computed; and the water amount of snowmelt and rainfall in the drainage basin area A of the n + 1 day snowmelt zone drainage basin and the water-recession flow of the river channel runoff in the drainage basin are combined, and the day runoff volume of the n + 1 day snowmelt zone drainage basin in the current snowmelt period is obtained by computing. According to the method and the device, the accuracy and the integrity of prediction of the day runoff volume of the snowmelt period can be improved.

The invention discloses a sewage sludge reduction dewatering method which is based on a reduction dewaterer. The reduction dewaterer comprises at least one dewatering module, wherein each dewatering module comprises a two-way cylinder (1) and two symmetrically arranged dewatering units (2); each dewatering unit (2) comprises a dewatering shell (2.1), a supporting plate (2.2) and an extrusion plate (2.3); the supporting plate (2.2) is positioned on the outer side of the dewatering shell (2.1); the extrusion plate (2.3) is positioned on the inner side of the dewatering shell (2.1); the dewatering shell (2.1) is provided with a feed port (2.4); the two-way cylinder (1) is provided with two piston rods (1.1); and the two piston rods (1.1) are respectively connected with the extrusion plates (2.3) of the two dewatering units (2). The feed dewatering and extrusion dewatering are carried out at the same time, thereby enhancing the work efficiency. Meanwhile, the feed pressure is ingeniously used for dewatering in the extrusion dewatering process, thereby enhancing the power use effect.

The invention discloses a method, a device and mobile equipment for realizing terminal-simulated mouse to operate equipment, and belongs to the field of Internet mobiles. The method comprises the following steps: acquiring sensor data collected by an internal sensor of a terminal, wherein the sensor data comprises acceleration data or acceleration data and angular velocity data; based on the sensor data, calculating the moved distance of the terminal in a scheduled time interval; based on the acceleration data, calculating the movement direction of the terminal in the scheduled time interval; transmitting the moved distance and the movement direction to the equipment connected with the terminal, in order to allow the equipment to move a mouse cursor according to the moved distance and the movement direction. The acceleration data and / or the angular velocity data are used for calculating the moved distance and the movement direction to be reported to the equipment, so as to realize cursor movement. Therefore, the effect of quick location and accurate location of the cursor during the terminal simulating mouse process is achieved.

The invention provides a development document generation method and device. The method comprises the following steps of: importing a control parameter in a preset program development management tool, wherein the control parameter is used for compiling parameter names of parameters appearing in each method in an API program into a class file corresponding to each method during program compilation; injecting a RequestMappingHandlerMapping to obtain related parameter information of each method in the API program, wherein the related parameter information comprises a URL, a Controller, a request parameter and a returned Header type; analyzing the request parameter to obtain a parameter name, a parameter type and parameter annotation content by using a reflective analysis technology; and packaging the related parameter information of each method in the API program and various analyzed contents to obtain an API development document. According to the method and device, development documents can be automatically generated, so that the problem that development document needs manual programming is solved.

The invention relates to a user behavior prediction method, which belongs to the field of computer technologies and solves a problem that the prediction result is inaccurate in the prior art. The userbehavior prediction method comprises the steps of constructing a t-moment behavior transfer probability matrix of a target user according to behavior data of a target user before the moment t; iteratively training a behavior prediction model of the target user based on the t-moment behavior transfer probability matrix and preset behavior influence factors; and predicting a next behavior of the target user based on the t-moment behavior transfer probability matrix, the preset behavior influence factors and a behavior prediction result of the target user at the previous moment according to thebehavior prediction model. According to the user behavior prediction method disclosed by the embodiment of the invention, behavior prediction is performed through combining related factors of user behaviors and a behavior transformation relation of the user, the behavior granularity of the user is refined, the user behaviors are comprehensively considered, model training is performed by combiningthe behavior transformation probability, and the prediction accuracy of the model is effectively improved.

The invention discloses a method for pushing the parking space information. The method comprises the steps of receiving the unoccupied parking space information reported by at least one first user terminal, wherein the unoccupied parking space information includes the parking lot identification information and the number information of unoccupied parking spaces around the location of the first user terminal; conducting the holistic approach on the unoccupied parking space information in the same parking lot according to the parking lot identification information; conducting the statistics on the number of unoccupied parking spaces in the same parking lot; generating the unoccupied parking space recommendation information of the parking lot; determining the position of the parking lot; selecting a plurality of second user terminals within a predetermined distance range around the parking lot; and pushing the unoccupied parking space recommendation information to the plurality of second user terminals. According to the technical scheme of the invention, the problem in the prior art that the unoccupied parking space information of a parking service system is not accurate can be effectively solved. Meanwhile, the real-time and accurate unoccupied parking space information can be pushed. Therefore, the trips of users can be facilitated.

The invention relates to a distributed temperature testing system and method for a hot dry rock deep well. The system comprises a distributed optical fiber temperature collection device, a logging winch, a well descending device and a temperature measuring optical cable, wherein the distributed optical fiber temperature collection device is used for completing collection of temperature along the temperature measuring optical cable and data processing; the logging winch is used for containing and transporting the temperature measuring optical cable and further used for providing a power source for fixed-depth and uniform-speed descending and winding of the temperature measuring optical cable; the well descending device is used for mechanical fixation and guidance for descending and winding of the temperature measuring optical cable; and the temperature measuring optical cable serves as a temperature sensing element and a data transmission channel. The system and the method achieve distributed temperature measurement of the hot dry rock deep well, and have the characteristics of resisting high temperature and high pressure, resisting electromagnetic interference, being free of limitation of severe environments like complex geometric space, and the like.

The invention discloses an urban traffic control sub-area division method based on fast global K-means spectral clustering, and relates the field of the urban traffic sub-area division. The basic thought thereof is taking traffic correlation between adjacent signal intersections in a road network as a division evidence, the method comprises the following steps: 1) selecting a to-be-researched roadnetwork to acquire a road topology structure and traffic flow data of the road network, and establishing a road network undirected graph; 2) computing a road network density Laplace matrix L and feature vectors corresponding to front k minimum feature values to construct a clustering feature matrix; 3) implementing the global K-means clustering algorithm on the clustering feature matrix X, and dividing the intersections in the same cluster into the same sub-area; and 4) judging whether the amount (absolute value of Vi) of the intersection in the sub-area after the clustering satisfies a constraint condition, performing boundary adjusting on the sub-area dissatisfying the constraint condition , and performing traffic control sub-area division on the road network G by adopting the fast global K-means spectral clustering. The time required by the clustering is shortened, and the traffic sub-area division efficiency can be improved.

The invention provides a rural load type load modeling method based on load curve decomposition. The method includes the following steps of decomposing a rural load curve, and calculating the proportion of each type of loads in all rural loads; calculating static load equivalent parameters, dynamic load equivalent parameters and power distribution network system impedance, combining the parameters and the impedance with power supply area network topology data of rural load nodes, and obtaining a rural load type load model. According to the rural load type load modeling method based on load curve decomposition, the defects of a traditional statistics and synthesis method are overcome, modeling can be rapidly, conveniently and accurately conducted on rural load stations, accuracy of power grid simulation calculation can be improved due to the load model produced through the method, and safe, reliable and economical running of a power grid is ensured.

A method for recommending an address of a pick-up point is provided. By acquiring the receiving address and according to the receiving address, the pick-up point information is determined by using theset pick-up point recommendation model, which is trained according to the historical pick-up address of the delivered express and the corresponding pick-up point information, and when the pick-up point information is determined, the pick-up point information is returned to the client. Thus, the original data deviation of the recommended pick-up point address is eliminated, and the beneficial effects of improving the accuracy of the recommendation result and the humanization of the recommendation content are obtained.

The invention relates to the field of computer vision, in particular to a binocular vision three-dimensional reconstruction method based on structured light and polarization. The specific technical scheme is as follows: a projector projects a sinusoidal fringe image generated by a computer onto a measured object; a left camera and a right camera acquire sine stripe images, which are modulated by the height, of the surface of the measured object and send the images to a computer; the computer carries out data processing on the sine stripe images containing the height modulation information, firstly solves a phase value through quality map guidance, then solves a preliminary point cloud coordinate of the surface of the measured object according to a geometrical relationship among the projector, the left camera, the right camera and the measured object, and calculates a point cloud coordinate of a high-reflection area based on a polarization method; and finally, the initial point cloud coordinates and the point cloud coordinates of the high-reflection area are registered to obtain complete point cloud to reconstruct a three-dimensional model of the measured object. The three-dimensional reconstruction method solves the problem that an existing three-dimensional reconstruction method based on structured light cannot well process reconstruction of a high-reflection area.

The invention discloses an acoustic design method for road noise barrier, which comprises the steps of: 1) importing a vector diagram with a Cad format for characterizing the terrains of noise barrier or grid maps with formats such as jpg, bmp, tif, and the like; 2) calibrating the imported graph to achieve the consistence of graphic proportion and actual proportion; 3) combining the actual situation to draw entities such as roads, terrains, houses, forests, grasslands, and the like based on the importing of the graph, and adopting a batch conversion function to finishing the basic work of acoustic modeling by adopting a batch conversion function if the graph is a Cad graph; 4) setting a noise prediction point at a protected sensitive point, calculating a noise value when no barrier exists, and determining an noise-reducing index of the barrier; 5) simulating the noise barrier, adjusting the height and the length of the barrier, and preliminarily calculating a geometrical dimension of the barrier which meets the requirement; and 6) re-calculating the noise-reducing effect of the barrier by combining the geometrical dimension of the sound barrier and simulating the barriers with different structures and sound absorption coefficients until the combination of the geometrical dimension, the structure and the sound absorption coefficient of the sound barrier which meets the requirement is calculated.

The invention discloses a fixed-point processing method and a fixed-point processing device. The method comprises the following steps of: respectively performing signal range analysis on all floating point signals according to all paralleling condition branches divided by a control-data flow diagram of a floating point algorithm to obtain the change range of all floating point signals and a smallest integer bit width, wherein the smallest integer bit width is a smallest positive integer value which is in accordance with the condition that a fixed-point signal does not overflow; respectively performing system precision analysis on all floating point signals subjected to signal range processing operation by using the change range to obtain a smallest decimal bit width of the floating point signals, wherein the smallest decimal bit width is in accordance with the condition that a system output precision error is less than or equal to a smallest positive integer value restricted by a predetermined error; and determining a result obtained by performing the fixed-point processing on the floating-point signals by using the smallest integer bit width and the smallest decimal bit width of the same floating point signal in all floating point signals. By the method, the accuracy of the fixed-point processing is improved.

The invention discloses a method and a device for aggregation of data in a datasheet. The method for aggregation of data in a datasheet comprises the steps of determining a first preset index sheet and a second preset index sheet; determining a first dimension main key, wherein the first dimension main key is a common dimension main key for the first preset index sheet and the second preset index sheet; associating the first preset index sheet with the second preset index sheet through the first dimension main key to obtain an associated datasheet; determining indexes to be aggregated in the second preset index sheet in the associated datasheet; collecting an aggregation result of the indexes to be aggregated corresponding to the first dimension main key, wherein the aggregation result of the indexes to be aggregated corresponding to the first dimension main key is a ratio of the sum of the indexes to be aggregated corresponding to the first dimension main key in the associated datasheet to the indexes corresponding to the first dimension main key in the first preset index sheet. Through the method and the device for aggregation of data in the datasheet, the problem that the aggregation computation result after association of two index sheets with dimension difference of 1 in the prior art is not accurate is solved.

The invention provides a load curve decomposition based load modeling method for a commercial load and a resident load. The method comprises the following steps: decomposing a load curve, calculating the proportion of each load in total quantity, and calculating a static load equivalent parameter, a dynamic load equivalent parameter and power distribution network system impedance. The invention provides the load curve decomposition based load modeling method for the commercial load and the resident load. The method overcomes the defects of a traditional statistics comprehensive method and can help to rapidly, conveniently and accurately model for a resident load station, by adopting a load model generated according to the method provided by the invention, the accuracy of simulating calculation of a power grid is increased, and the safe, reliable and economical operation of the power grid is guaranteed.