Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

48results about How to "Rich scene" patented technology

Text detection method of document image in natural scene

The invention discloses a text detection method of a document image in a natural scene. Commonly used Chinese characters are selected to make Chinese character pictures, a dataset 1 is formed, randomrevolving and cropping operations are carried out on labeled document images, then a manner of Poisson cloning is used to fuse different background images, and a dataset 2 is formed; the dataset 1 isadopted to carry out training of a text classification model on a VGG16 network, and after the model converges, obtained parameters are used to initialize a fully convolutional neural network model, and the dataset 2 is used to train the model; the trained fully-convolutional neural network model is used to process the image, a classification situation of each pixel point is obtained according toa maximum probability method, and a text-non-text binary image is formed; a method of connected regions is used to obtain text regions, the original image is binarized, and only text information in the text regions in the text-non-text region binary image is extracted to obtain a text binary image; the image is corrected through a maximum variance method; and projection is carried out again on thecorrected image, and the text-non-text region binary image is refined.
Owner:BEIJING UNIV OF TECH

Data mining method of terminal user social network, correlation method, device and system

The embodiment of the invention discloses a data mining method of a terminal user social network, a correlation method, a device and a system, which are characterized in that the user identification of a terminal is obtained; then, on one hand, communication records of a user are obtained by utilizing the user identification, and first social information is obtained according to the user identification and the communication records; on the other hand, a data packet of social network service visited by the user is obtained according to the user identification, and second social information is obtained through the user identification and the data packet; on the third hand, information distributed by the user on an internet can also be obtained according to the user identification, and then, the third social information is obtained according to the user identification and the distributed information; and finally, a user social database is established or updated by utilizing the first social information, the second social information and the third social information to form the social network with rich scenes, detailed contents and complete information, thereby conveniently improving the efficiency of various services required to be operated based on the social network.
Owner:HUAWEI TECH CO LTD

Automatic generating method and apparatus for RPG game scene path

The invention discloses a method and device of path automatic generation of 2D / 3D RPC game scenes in mobile platform. The method includes following steps: initialize 3D scene model, plan paths, generate main paths and generate branch paths etc. The device includes following modules: path planning module, main path module, branch path module, loop path module, obstacle module and pre-processing module. The invention simplifies 3D scenes to be 2D map of terrain. Then, the map can be represented as 2D paths.
Owner:SUN YAT SEN UNIV

Virtual upper limb control system based on myoelectricity and motion capture and method of system

The invention discloses a virtual upper limb control system based on myoelectricity and motion capture and a method of the system. The system comprises a myoelectric device, a motion capture device, avirtual game development environment and a virtual/augmented reality device, wherein the virtual game development environment comprises a virtual upper limb model and a virtual game scene; the myoelectric device is used for detecting myoelectric signals and gesture information of different gestures and sending the myoelectric signals and the gesture information to the virtual upper limb model; the motion capture device is used for acquiring joint coordinates and spatial posture information of upper limbs and sending the joint coordinates and the spatial posture information to the virtual upper limb model. The virtual game scene receives data sent by the virtual upper limb model and sends the data to the virtual/augmented reality device, the virtual/augmented reality device feeds back visual auditory stimulation to the user, and virtual objects are placed in a corresponding real environment for operation. The system has the advantages of being low in cost, high in precision, comfort degree, timeliness and interactivity and diversified in scene, has an upper limb integrated game pattern and can increase the game enthusiasm of users.
Owner:深圳睿瀚医疗科技有限公司

Simulation management method, device and equipment for railway vehicle service receiving and dispatching service

The embodiment of the invention discloses a simulation management method, device and equipment for railway vehicle service receiving and dispatching service. The simulation management method comprises: constructing a railway vehicle service receiving and dispatching service simulation library according to given station yard graph data, interlocking table data, running chart data and receiving anddispatching operation standard data; obtaining identity information of a target operator; selecting a target simulation operation test question corresponding to the identity information of the targetoperator from the railway train service receiving and dispatching service simulation library; providing the target simulation operation test question for the target operator, and obtaining operation information data of the target operator for the target simulation operation test question; and obtaining evaluation information of the target operator according to the operation information data and standard operation data of the target simulation operation test question. According to the invention, analog simulation can be carried out on railway vehicle service receiving and dispatching services,the application range is wide, the scenes are multiple, the exercise content is rich, and the human intervention is little.
Owner:沈阳风驰软件股份有限公司

Urban rail transit full-scene passenger flow prediction method and system

The invention belongs to the technical field of rail transit, and relates to an urban rail transit full-scene passenger flow prediction method and system. The current passenger flow analysis technology cannot provide all-around support for urban rail transit operation organization. The invention provides the urban rail transit full-scene passenger flow prediction method, which comprises the following steps: analyzing the travel characteristic rule of a single passenger in an urban rail transit system, and analyzing the influence of external conditions on the individual travel behavior of the passenger; monitoring whole-network passenger flow distribution in real time, starting from a time dimension, realizing a full-scene urban rail transit passenger flow prediction technology, giving an alarm prompt under an abnormal condition, and providing passenger flow data support for passenger flow organization adjustment or driving organization adjustment for emergency disposal; generating the corresponding statistical indexes by taking landmarks compiled by metro and the recently developed national standard of Urban Railway Traffic Operation Index System as references, and supporting user-defined report query and data export. Reference is provided for urban rail transit passenger flow control and train operation scheduling.
Owner:BEIJING JIAOTONG UNIV

Omni-channel chain operation background system for quick response

The invention discloses an omni-channel chain operation background system for quick response, which comprises an intelligent middle platform, a cashier terminal, a database, a server, an updating module and an intelligent middle platform coupon cloud platform, and is characterized in that the intelligent middle platform is a unified platform oriented to commercial ecology and is used for releasingthe association relationship between each ERP and CRM of the background and the front end; the cashier terminal is used for customer consumption and terminal data management; the server side can be in butt joint with a plurality of databases such as Oracle, DB2, Sqlserver and Mysql to realize data transmission; the system has the beneficial effects that a powerful promotion model and an operationengine support most promotion modes in the retail industry, the business matching degree is high, the adaptability is high, and the operation performance is good; a perfect coupon model and operationand pre-calculation engine supports most coupon marketing scenes on the market, and is high in adaptability, rich in scene and good in operation performance; and different types of interface butt joint modes are provided, so that different systems can be integrated efficiently.
Owner:深圳市昂捷信息技术股份有限公司

Database self-learning optimization method and device based on flow mirror image

ActiveCN110134665ADouble speed adjustableSelf-learning optimization time shortenedDatabase management systemsSpecial data processing applicationsAdaptive learningSelf training
The invention provides a database self-learning optimization method based on a flow mirror image, and the method comprises the steps: obtaining a latest database configuration parameter of a to-be-optimized database through a self-adaptive learning method, and sending the latest database configuration parameter to a training database; establishing a database mirror image model consistent with thedatabase model of the database to be optimized, and sending the database mirror image model to the training database; establishing a flow mirror image model consistent with the flow model received bythe to-be-optimized database, and sending the flow mirror image model to the training database; obtaining a training result fed back to the self-training database; updating database configuration parameters of the to-be-optimized database according to a training result of the training database; and repeating the above steps for R times, obtaining the latest database configuration parameter when the TPS throughput of the training database in the R training results is the maximum value, and configuring the latest database configuration parameter as the final database configuration parameter to the to-be-optimized database. For a real traffic scene, the performance of the database can be effectively improved.
Owner:BEIJING BAIDU NETCOM SCI & TECH CO LTD

Intelligent cloud platform for pedestrian analysis

The invention discloses an intelligent cloud platform for pedestrian analysis. The intelligent cloud platform comprises a face analysis engine which is used for carrying out the analysis of the identity characteristics and attribute characteristics of a face, a pedestrian analysis engine which is used for analyzing the attribute characteristics of each part of the pedestrian, a hybrid analysis engine which is used for performing unified analysis modeling and model storage on the face and pedestrian data under the characteristic space-time scene, a training engine which is used for continuouslyupdating the model performance of the cloud platform, and the training input data is in an automatic and manual mode, and a cloud computing architecture which is used for an SaaS architecture and anarchitecture of a software engineering part of data caching, storage and request agency. For the intelligent cloud platform for pedestrian analysis, the face recognition technology and the pedestrianre-recognition technology are combined; far and near pedestrians in a scene are detected, recognized and structurally analyzed; and the integrated identity recognition based on the face and pedestriancharacteristics can improve the accuracy of pedestrian recognition and can better adapt to richer scenes and user requirements.
Owner:苏州书客贝塔软件科技有限公司

Flood-proof household safety monitoring system with multiple emergency measures

The invention discloses a flood-proof household safety monitoring system with multiple emergency measures. The flood-proof household safety monitoring system comprises a household safety monitoring system server, a central node, a routing node, a water sensor node, a set top box, a user terminal and a broadcasting device, wherein the water sensor node indirectly communicates with the central node by the routing node in a wireless way or the water sensor node directly communicates with the central node; the central node receives the information from the water sensor node and then uploads the information to the set top box; the set top box transmits the information to the household safety monitoring system server by an internet; after the household safety monitoring system server receives the information, a short message is sent to a user and informs the user that flood occurs in the house and the user needs to check in time; an alarming signal is given by the broadcasting device, people in the house can find the flood as soon as possible; an instruction is sent to close a water valve. The user can find the flood in time, the economic and personnel loss is avoided, the water resource waste caused by the flood is avoided, the scene in the house can be more perfectly monitored by the household safety monitoring, and the requirements of the modern people on the household environment monitoring can be more conformed to.
Owner:QINGDAO PENGHAI SOFT CO LTD

Video connection method and device

The invention relates to a video connection method and device, and the method comprises the steps: obtaining a video connection request set at a preset matching moment, carrying out the pairwise matching of video connection requests in the video connection request set, and obtaining at least one group of matched video connection requests, creating a video connection room corresponding to each group of mutually matched video connection requests, generating identification information of the video connection rooms, and respectively sending the identification information of the video connection rooms to clients corresponding to the mutually matched video connection requests corresponding to the video connection rooms, and sending the identification information of the video connection room to the corresponding client, so that the corresponding client enters the video connection room for video connection according to the identification information of the video connection room and acquires public information of the opposite side during video connection. By a mode of matching the video connection request, the client performs video connection in a one-to-one video connection scene, so thatthe privacy and the adaptability of video social contact can be improved.
Owner:BEIJING DAJIA INTERNET INFORMATION TECH CO LTD

Underwater image real-time restoration model based on self-attention mechanism and GAN

The invention relates to an underwater image real-time restoration model based on a self-attention mechanism and a GAN, and belongs to the technical field of deep learning. A generative adversarial network is used as a basic framework in the invention, the generative network adopts a coding-decoding structure, a synthetic image is generated at an original resolution through nine residual blocks and deconvolution operation, a self-attention module is introduced, richer advanced features can be captured to improve model performance, and in order to keep image content and remove underwater noise at the same time, the discrimination network adopts a structure of a multi-branch discriminator comprising an adversarial branch and a criticality branch. The self-attention module is added, the model structure is improved, the problem of low time efficiency of underwater image processing is solved ideally, and the method based on GAN-RS is trained to adapt to various underwater conditions, and has excellent real-time processing performance.
Owner:GUILIN UNIV OF ELECTRONIC TECH +1

Training data generation method and device, computer readable medium and electronic equipment

The invention provides a training data generation method, a training data generation device, a computer readable medium and electronic equipment, and relates to the technical field of image processing. The method comprises the following steps: acquiring a video set, wherein the video set comprises at least one 2D video; performing frame screening on each 2D video based on a preset tracking condition to obtain a tracking frame sequence corresponding to each 2D video; and performing three-dimensional reconstruction based on each tracking frame sequence to obtain a reconstructed image and depth data corresponding to the reconstructed image, and taking the reconstructed image and the depth data corresponding to the reconstructed image as first training data. According to the method and the device, videos which exist on the Internet, are shot by different devices and are rich in scenes can be converted into deep truth values, and training data with a large quantity and rich scenes can be generated, so that machine learning or deep learning networks related to the deep data can be trained based on the data.
Owner:GUANGDONG OPPO MOBILE TELECOMM CORP LTD

Wire twisting device for elevator traction steel wire rope production

The invention discloses a wire twisting device for elevator traction steel wire rope production, and relates to the technical field of elevator production and manufacturing. The wire twisting device for elevator traction steel wire rope production comprises a workbench, wherein a vibration aging mechanism, a lubricating mechanism, a wire twisting mechanism, a translation mechanism and a transmission mechanism are arranged on the workbench, a threaded rod is rotationally installed on the outer surface of the top of the workbench, a limiting disc is fixedly installed on the outer wall of the threaded rod, three sets of steel wire rollers and a set of fixing discs are installed on the outer wall of the threaded rod in a threaded mode, the three sets of steel wire rollers are attached to one another, and the fixing disc is located above the three sets of steel wire rollers. According to the wire twisting device for elevator traction steel wire rope production, stress left on a steel wire is reduced in a vibration mode before wire twisting operation, so that the steel wire is not prone to rebound and deformation in subsequent wire twisting operation, the wire twisting quality of the device is guaranteed, compared with a method for reducing stress by heating a workpiece, the device has the beneficial effects of being efficient, capable of saving energy and safe, and convenient popularization and application is facilitated.
Owner:湖州和通电梯部件有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products