191results about How to "Get it quickly and accurately" patented technology

The invention discloses a three-stage convolutional neural network-based object grabbing detection method. The method comprises the following steps of: 1, obtaining a data set; 2, constructing networkstructures of first-stage, second-stage and third-stage convolutional neural networks, and training the convolutional neural networks; 3, obtaining pre-selected grabbing boxes of a target object andjudging values of the pre-selected grabbing boxes by utilizing the trained three-stage serial convolutional neural network; 4, obtaining an optimal grabbing box through the judging values; and 5, determining a position and a posture of the target object. The method is capable of improving the correctness of the grabbing boxes and realizing high-correctness grabbing for unknown objects.

The invention discloses an air pollution monitoring system based on Beidou positioning. The air pollution monitoring system comprises a mobile mechanism, a detection unit and a data processing unit, wherein the mobile mechanism is used for moving and comprises a fly lifting feed mechanism and a direction control unit; the detection unit is arranged on the mobile mechanism, is used for detecting air pollution indexes, comprises inhalable particles, chemical components and bacterium concentration, and is used for detecting the air direction and the air speed; the data processing unit is used for receiving the position information of the mobile mechanism, the air pollution indexes, the air direction and the air speed, used for generating an air pollution distribution diagram according to the air pollution indexes and the position information of the mobile mechanism, used for estimating the position of a pollution source according to the air pollution distribution diagram, the air direction and the air speed, and used for dividing pollution areas according to the preset air pollution concentration thresholds. Therefore, high-precision and high-reliability pollution positioning can be achieved, the air pollution distribution situation can be monitored in real time, and the pollution resources can be estimated according to the air pollution distribution diagram, the air direction and the air speed.

The invention relates to a service network station matching method and system. The method comprises the following steps of: obtaining the position of a customer; matching the position with a service area of each to-be-selected network station, wherein the service area is a geographic area formed by a plurality of service boundary points of the to-be-selected network station; and when the position is located in the service area of one of the to-be-selected network stations, determining the to-be-selected network station as a service network station of the customer. According to the method and system, geographic position matching is carried out on the longitudes and latitudes of customers and the service areas, the problem of matching failure caused by nonstandard address information or incomplete service network station information for matching when only address information obtained by text input is matched is avoided, and the success rate of matching between the customers and the service network stations is effectively improved.

The invention provides an automatic microscope focusing method, comprising the steps of utilizing a liquid lens of which the focal length is capable of being controlled by voltage in a microscope optical path to be used as a zooming element, capturing a video stream while performing continuous stepping changing on the driving voltage of the liquid lens, performing focusing evaluation on each captured image, utilizing a focal point searching method to find the liquid lens voltage, positioned at the optimal focusing state, of the microscopic image, and then driving the liquid lens voltage to such a position to finish the automatic focusing. The automatic microscope focusing method fully utilizes the advantages of fast responding speed and small delaying of the liquid lens and solves the problems of slow and inaccurate focusing during the automatic focusing process by using a traditional mobile lifting table.

The invention belongs to the technical field of metal corrosion morphology grade judgment, and discloses a corrosion grade information processing method and system based on image recognition, and themethod comprises the steps: obtaining a labeled laboratory metal morphology picture as a data set; carrying out the de-duplication on the obtained training data set, randomly selecting a half of images of each category as a test set, and taking other images as training set images; expanding the training data set, namely carrying out the data enhancement; building a convolutional neural network model by using a Keras framework, carrying out model training, and storing optimal model parameters; and analyzing the grade of the corrosion morphology of the metal surface by utilizing the trained optimal model. According to the method, the judgment of the metal corrosion morphology grade is realized by using a computer vision technology, so the judgment efficiency is greatly improved and the judgment uniformity and objectivity are ensured under the condition of ensuring the judgment accuracy. The method can assist engineers engaged in material research in judging the metal corrosion grade, andvarious difficulties of a previous judgment method are solved.

The invention discloses a driving state recognition method based on approximate entropy template matching, and the method comprises the steps: 1, sample library building, wherein samples of one type in the sample library are a plurality of steering wheel turning angle signals in a normal driving state, and samples of the other type in the sample library are a plurality of steering wheel turning angle signals in a dangerous driving state; 2, road information segmentation based on the approximate entropy template matching: carrying out the call of a signal correction module based on the approximate entropy template matching to correct the steering wheel turning angle signals in the library sample, wherein the correction process of any one steering wheel turning angle signal is as follows: carrying out the EMD (Empirical Mode Decomposition), the effectiveness recognition of an intrinsic mode function component, and signal reconstruction; 3, feature extraction; 4, two-class model building and training; 4, driving state information collection and synchronous classification. The method is simple in steps, is reasonable in design, is easy and convenient to implement, is good in use effect, can accurately recognize the driving state of a driver simply and conveniently, and is high in recognition precision.

The invention relates to an image acquisition method and an electronic device using the image acquisition method. The image acquisition method is applied in the electronic device with an image acquisition unit; the image acquisition method comprises following steps: acquiring multiple first images of a photographed scene; sensing and recording the attitude information data of the electronic when acquiring each image in the multiple first images; and splicing into a second image based on the pose information data and the multiple first images.

The invention discloses an indoor multi-target tracking method using a density-based fast search clustering algorithm, and mainly solves the problems that the present indoor positioning technology has bad precision and cannot track multiple targets. The technical scheme comprises the steps of: 1) obtaining wireless link information and performing pre-treatment, and obtaining a fluctuation link set; 2) removing outliers in fluctuation link intersection points; 3) clustering normal points and obtaining the number and positions of the targets; 4) correcting the target number by using a Hidden Markov Model; 5) correcting the target positions by using fuzzy C clustering; 6) joining the number and position information of the targets and generating target moving tracks; and 7) performing multi-particle filtering correction on the target moving tracks and realizing accurate tracking on the targets. The indoor multi-target tracking method using the density-based fast search clustering algorithm reduces the influence of environment on the positioning precision, and improves the robustness. The precision can reach a centimeter level. The indoor multi-target tracking method using the density-based fast search clustering algorithm can be used for the indoor multi-target tracking and monitor region security.

The invention is suitable for the technical field of environment monitoring, and provides an environment monitoring method, system and platform and a storage medium, and the method comprises the steps of obtaining corresponding environment data, collected by a sensor system, in a monitoring region where the sensor system is located; inputting the environment data into a trained alarm model to obtain alarm information; and if the alarm information satisfies an early warning condition, sending the alarm information to a client corresponding to the monitoring area. According to the environment monitoring method, when the sensor system collects the corresponding environment data in the monitoring area and the obtained alarm information meets the early warning condition, the alarm information is sent to the client side corresponding to the monitoring area where the sensor system is located, so that when the pollution degree of the monitoring area is heavy, a worker can take corresponding measures in time.

The embodiment of the application discloses a voice data processing method and device, and a storage medium. The voice data processing method comprises the following steps: obtaining target text dataassociated with target voice data; determining a title text in the target text data based on segmented keywords in the target text data; obtaining service auxiliary data matching the subject information of the target text data, and obtaining an auxiliary text associated with the title text in the service auxiliary data; and determining the target text data as outline text data according to the auxiliary text. By using the application, the text outline can be generated quickly and accurately, so as to improve the generation efficiency of the text outline.

The invention discloses a remote sensing detection and spatial hot area identification system and method for a contaminated site. The system comprises a spatial data acquisition and preprocessing module, a contaminated site remote sensing detection module and a contaminated site hot area detection module. The system serves the work of contaminated site monitoring, hot area detection and the like.The system comprises a novel contaminated site remote sensing image detection method framework, and an identification method for a centralized distribution area of the contaminated site is realized. The system has the characteristic of high practicability, and has urgent application requirements in the related fields of current contaminated site situation monitoring, contaminated site planning, ecological environment restoration and the like.

The invention relates to a mobile consumer service information system, which comprises a client server platform, a local facilitator network point, and a mobile maintain terminal, which are connected via network, wherein said local facilitator network point and the mobile maintain terminal, via wireless network, share the information; the engineer of mobile maintenance has mobile terminal feedback device, to feedback the consumer information to the client server platform; and the client server platform processes the received information, to send it to the facilitator automatic distribute network point, which will distribute server to the engineer. The invention can improve the on-time property of service, improve the accuracy and improve the service quality.

The invention discloses a binocular vision-based method for detecting a barrier in an indoor shadow environment. A binocular vision system is adopted. The method comprises the steps of (1) removing a shadow, extracting color saturation of an original image, performing fusion with image information subjected to shadow removal, and adaptively adjusting and increasing grayscale difference of the ground and the barrier through environmental brightness information; (2) filling a ground region of a fused image by utilizing a seed filling algorithm, and obtaining a barrier region through threshold segmentation, corrosion and expansion operations; and (3) performing matching in a right camera image by taking the obtained barrier region as a template through applying the binocular vision system to calculate parallax of a central point, and calculating three-dimensional coordinates of the central point and the width and distance of the barrier. The invention furthermore discloses a binocular vision-based apparatus for detecting the barrier in the indoor shadow environment. According to the method and the apparatus, the barrier can be completely extracted in the indoor shadow environment; and the method and the apparatus are simple and efficient, have relatively good real-time property and relatively high precision, and are suitable for mobile robot navigation and barrier avoidance.

The invention provides an electronic tag based object identification method, device and system. The electronic tag based object identification method comprises the steps of setting corresponding electronic tags on objects to be identified; setting a detection device capable of identifying the electronic tags and a transmitting device used for transmitting detection signals of the electronic tags in an intelligent bag; when any one object is placed in the intelligent bag, detecting the electronic tag of the object by the detection device in the intelligent bag, and generating a corresponding detection signal; and transmitting the detection signal to a mobile terminal associated with the intelligent bag by the transmitting device in the intelligent bag, wherein the mobile terminal parses the detection signal to obtain the electronic tag, and identifying the objet corresponding to the electronic tag. According to the object identification method provided by the embodiment of the invention, the detection device can perform detection quickly as long as any one object is placed in the intelligent bag, thereby ensuring timely acquisition and update for the object information in the mobile terminal.

The invention provides a deep learning data sharing method based on a block chain mode, and a storage medium. The method comprises the steps of storing deep learning data and a target modeling model in all nodes of a block chain network, wherein the deep learning data is shared in the block chain network, and the target modeling model is protected; splitting the deep learning data into metadata with corresponding data attribute tags; when a deep learning request corresponding to the target modeling model is received, determining the corresponding data attribute tags according to the target modeling model, and obtaining the corresponding metadata; and according to the corresponding metadata, extracting the deep learning data corresponding to the target modeling model to perform learning. The method has the advantages that a deep learning mode is optimized, and one step of reinforcement learning can be achieved, so that the learning efficiency can be remarkably improved; and furthermore,the learning data can be reused, and the learning cost and time can be saved.

The invention discloses an Ethernet-based network control system. The system comprises a server, a network control terminal and a distributed network node, wherein the server is used for responding to a network control request of a network control terminal and transmitting a network control data transmission command to the distributed network node; the network control terminal is used for transmitting the network control request to the server and receiving network control data from the server; the distributed network node is used for transmitting the network control data to the server after receiving the network control data transmission command of the server; the distributed network node comprises field equipment and a field controller, wherein the field controller is connected with the field equipment and is provided with a Windows operating system; and the server, the network control terminal and the field controller are all connected with the Ethernet. The Ethernet-based network control system can realize the function of acquiring the network control data intuitively, accurately and rapidly under the conditions of saving bridging equipment, a network conversion protocol and a driver, and has strong system configuration flexibility and expansion capacity.

The invention provides a method and a device for generating and reading dynamic two-dimensional codes. The method for generating the dynamic two-dimensional codes comprises the following steps: encoding to-be-encoded information into information code words; calculating distribution centers of the information code words in a code word space according to the distribution of the information code words in the code word space; carrying out redundancy encoding on the encoded information cord words according to the distances from the information code words to the distribution centers to form finally encoded information cord words; and generating a dynamic two-dimensional code image sequence including a preset number of two-dimensional code images according to the encoded information cord words, and carrying the values of the distribution centers in the code word space and the initial sum in each frame of two-dimensional code image. Through application of the method and the device, original to-be-encoded information can be quickly and accurately obtained from the dynamic two-dimensional codes even if the scanning frequency of reading equipment is lower than the display frequency of the dynamic two-dimensional codes.

The embodiment of the invention discloses a data processing method and device, equipment and a storage medium, and the method comprises the steps: obtaining a text vector corresponding to an atlas text in a first knowledge atlas, and obtaining a text vector corresponding to an atlas text in a second knowledge atlas; in the first knowledge graph and the second knowledge graph, determining the textvectors under the same graph text dimension as a same-dimension vector set, and performing clustering processing on the text vectors in the same-dimension vector set to obtain a to-be-aligned oligomerization cluster; selecting a target cross-map vector pair from the to-be-aligned clustering cluster, wherein the target cross-map vector pair comprises a text vector belonging to a first knowledge mapand a text vector belonging to a second knowledge map; and determining an alignment result between the first knowledge graph and the second knowledge graph for the graph text according to the targetcross-graph vector pair. By adopting the embodiment of the invention, the alignment efficiency and the alignment accuracy of the knowledge graph can be improved.

The invention discloses a method, a server and a system for achieving a shared file. A virtual disk is created according to a file list structure of a file allocation table FAT32 adopting 32 bits, dispersed files are organized in the same virtual disk, samba share is added through operation in one time so that a shared resource seen by a client is a disk partition, and the files seen by the client are controlled through a file list and the server. Besides, by modifying the file list, the files and folders in the virtual disk and the corresponding relations between virtual files and real files are controlled conveniently, structures of the folders and the files of the client are organized, and the files capable of being read by the client are controlled. Accordingly, the shared file can be acquired rapidly and correctly with low consumption through the method, the server and the system.

The invention provides a commuting house searching method and device. The method comprises the following steps that: dividing a map into geohash blocks; calculating corresponding commuting time when different commuting ways between the geohash blocks are used for commuting, and storing the commuting ways among a finite number of geohash blocks and corresponding commuting time in a first database;obtaining the coordinate of a user and the acceptable commuting time of a user, and converting the coordinate of the user into the corresponding geohash block, inquiring the first database according to the converted geohash block and the acceptable commuting time of the user, and obtaining each target geohash block which meets commuting time requirements and a commuting way and commuting time independently corresponding to each target geohash block; and according to each target geohash block, independently inquiring the first database, obtaining a housing estate or a housing estate set independently corresponding to each target geohash block, and displaying each obtained housing estate or housing estate set and the corresponding commuting time and commuting way. By use of the method, housing resources which meet the commuting requirement can be quickly and accurately obtained.

The invention provides a solar power supply greenhouse, and belongs to the field of greenhouses. The solar power supply greenhouse comprises a greenhouse frame, film, a biological lamp, sensors, a film rolling machine, an opening and closing mechanism, solar cell panels, movable supports, storage cells and a controller; the film is arranged on the greenhouse frame and connected with the film rolling machine, the top of the greenhouse frame is provided with a ventilation opening, and the ventilation opening is provided with the opening and closing mechanism; the solar cell panels are arranged on the two sides of the top of the greenhouse frame, movably connected with the greenhouse frame through the movable supports and connected with the storage cells through adapters; the controller comprises a collection module, a processing module, a time module, a setting module, a control module and a display module. According to the solar power supply greenhouse, the controller can acquire the temperature and the humidity inside the greenhouse through the temperature sensor and the humidity sensor; the solar cell panels and the storage cells can effectively utilize solar energy, surplus electric energy is stored and reutilized by the storage cell, and the vegetable planting cost is reduced.

The invention discloses an unmanned aerial vehicle with an air detection function. The unmanned aerial vehicle comprises an unmanned aerial vehicle body, a detection box and an MCU controller. The detection box and a main vehicle box are installed in an inner cavity of the unmanned aerial vehicle body. A PM2.5 sensor, a temperature sensor and a humidity sensor are arranged at the top of an inner cavity of the detection box sequentially from left to right. A carbon dioxide sensor, a sulfur dioxide sensor and a nitrogen dioxide sensor are arranged at the bottom of the inner cavity of the detection box sequentially from left to right. The input end of the MCU controller is electrically connected with the PM2.5 sensor, the temperature sensor, the humidity sensor, the carbon dioxide sensor, the sulfur dioxide sensor, the nitrogen dioxide sensor, a GPS locating module, a ranging device and the output end of a camera. The output end of the MCU controller is electrically connected with a gas inlet valve, a gas outlet valve, a flight controller and the input end of a speed governing controller. Through the unmanned aerial vehicle, the air quality condition in a high altitude area can be obtained, detection data is obtained accurately, and the safety guarantee capability is high.

The invention discloses a cache-based quick query method. The method comprises the steps of obtaining a query demand of a user for querying required information; detecting whether required data existsin cache data or not on the basis of the query demand; if the required data exists, acquiring the required data, and ending the query; if the required data does not exist, switching to a database toinquire required information, and calling after the required information is inquired; and taking the called required information data as value, taking the id of a class name plus the required data askey, and storing the value and the key in cache data. The consistency of data in a database and data in a cache is guaranteed through distributed cache middleware in a cluster environment, high availability and persistence of the cache middleware are guaranteed, the problem that the disk response speed is low due to the fact that the data size is too large is solved, and data needed by a query command are rapidly and accurately obtained; the cache of the memory space is periodically cleared, and the situation of overflow due to overlarge data volume is prevented.

The invention discloses a mangrove forest remote sensing rapid and accurate extraction method based on a cloud platform, and relates to the field of mangrove forest distribution diagram manufacturing, and the method comprises the steps: firstly, screening Sentinel-2 images with good quality through a Google EarthEngine cloud platform, and forming an image set; calculating a vegetation index NDVI of each image in the image set to obtain an NDVI image in space; forming a low-tide image formed by the maximum NDVI pixels; erasing an area without vegetation by using a mask function; utilizing an MFVI index to identify mangrove forest space distribution information; and erasing a non-mangrove forest region by using a mask function to obtain a spatial distribution diagram of the mangrove forest. According to the method, the difficulty of mangrove forest remote sensing interpretation caused by periodic tidal water submerging is overcome, and the problems that the submerged mangrove forest is missed to be painted and inland vegetation is wrongly divided into mangrove forests are solved. The mangrove forest interpretation method is fast and effective, improves the precision and credibility of mangrove forest interpretation, and has repeatability and robustness.

The invention provides a method for quickly searching tree structure data in a database and a storage medium. The method comprises the steps: a node path corresponding to the tree structure of each node is stored to a database table; a request to search all nodes in the tree structure including a node field under a node is received; a node containing the node field is obtained from the database table by searching, and a corresponding node path of the node contains the node. The invention can not only realize quick search for accurate results, but also greatly reduce operation complexity. Operation is only based on the database table without extra memory overhead, which significantly reduces the space complexity. Furthermore, support can be realized only by adding one field in the databasetable, and the operation is simple and practical. The invention is particularly suitable for fast and accurate searching of a large amount of tree structure data, and therefore has a wide applicationprospect and further improves the practicability.

The invention provides a variable-refrigerant-flow protocol conversion control method, device and system and an air conditioning system. The method comprises the following steps of receiving inner unit signals from inner units, and determining an inner unit communication protocol executed by the inner unit signals; when the inner unit communication protocol is different from an outer unit communication protocol, converting the communication protocol executed by the inner unit signals into the outer unit communication protocol; sending the inner unit signals converted through the communicationprotocol to variable refrigerant flow outer units. By means of the variable-refrigerant-flow protocol conversion control method device and system and the air conditioning system, the universality of the variable-refrigerant-flow air conditioning system can be improved, and the development efficiency and the using effect on the air conditioning system can also be improved.

The invention relates to a fault diagnosis method based on a self-adaptive optimized pulse neuromembrane system. The method includes the steps of obtaining action information of a relay protection device and trip operation information of a circuit breaker from a monitoring and data acquisition system of a power grid; according to the acquired information, identifying a power outage area and determining suspicious faulty components in the power outage area; according to a logical relationship among the suspicious faulty components, protection actions and circuit breaker trip operations, expressing a fault diagnosis problem as a 0-1 integer programming problem; increasing a learning rate in an original optimized pulse neuromembrane system, introducing intersection and variation to obtain theself-adaptive optimized pulse neuromembrane system, solving an objective function and then determining a faulty component; establishing a human-machine interactive interface window and displaying a fault area, the faulty component, the protection actions and the maloperation and operation failure conditions of the circuit breaker in real time. According to the method, the correctness of fault diagnosis is high, the assistance of other heuristic optimization algorithms is not needed, the iteration frequency can be greatly reduced, and an optimal value is found quickly.

The embodiment of the invention provides an information recommendation method and an information recommendation device, which relate to the technical field of computer application. According to the technical scheme provided by the embodiment of the invention, when a user is unfamiliar with a goods in a goods advertisement or a picture, one-touch scanning is triggered through a one-touch scanning function of an application in user equipment to acquire a goods information request, the user equipment can trigger a camera to scan and acquire the goods picture; the goods picture is subjected to picture processing to extract goods feature information in the picture; and according to the goods feature information, the goods information matched with the goods feature information is recommended. Thus, the recommendation information of an unfamiliar goods can be acquired quickly and accurately, and the experience of the user is greatly improved.

The invention discloses a metro automatic management system based on Beidou positioning. The metro automatic management system comprises a data collecting module, a data processing module and a driving module, wherein the data collecting module comprises a position sensor, infrared emitters, infrared receivers and pressure sensors; the position sensor is used for sending a request of obtaining train position coordinates; the infrared emitters and the infrared receivers are arranged on the first side wall and the second side wall of the lower part of a platform in pairs, and are used for measuring railway safety parameters; each pressure sensor is arranged on a railway, and is used for measuring and sending pressure values and the pressure sensor number; the data processing module is used for receiving the train position coordinates, the railway safety parameters, the pressure values, the time and the pressure sensor number, and sending control signals; the data processing module comprises a starting and stopping controller and a speed controller; and the driving module controls the acceleration, the deceleration, the stopping or the running starting of a metro train. Therefore the metro automatic management system has the advantages that the automatic control of the metro train can be realized; and the automatic recognition and disposal on passengers falling off from the platform can be realized.

The invention discloses a graphical office room management system based on Web. The system can graphically display the structure (2D / 3D), area, personnel, unit and other information of a house. The system is characterized in that the system also includes: a toolbar, a canvas, an information window, legend, 3D display, etc. The toolbar has the functions of moving, drawing solid lines, drawing dotted lines, drawing rectangles, adding characters, adding doors, adding stairs, adding elevators, copying, pasting, uploading accessories and the like. The invention further comprises a graphical officeroom management method based on Web. Graphical management of office rooms of party and government organs and enterprises and public institutions is realized; visual editing and storage of data such asthe structure, the use type, the unit, the personnel, the type and the record of the office room are achieved, many defects of original office room management are overcome, the office room managementis simplified, various data of various office rooms can be rapidly and accurately obtained, and manpower, resources and time are saved.