152results about How to "Achieve understanding" patented technology

A data display device for displaying, on a display unit, a group of images related to a plurality of pieces of file data comprises: an operation unit operable to receive, from a user, a display operation pertaining to the group of images related to the pieces of file data; a storage unit storing cluster information on clusters that are generated by grouping the pieces of file data according to a grouping condition; and a control unit operable to display, on the display unit, the group of images related to the pieces of file data so that the user can distinguish between clusters, in accordance with the display operation received by the operation unit and with the cluster information stored in the storage unit.

The invention provides a monitoring system and a monitoring method for monitoring performance of a cloud host. The monitoring method comprises the following steps: collecting a virtual resource utilization state of the cloud host by simultaneously using a cloud host Agent module, a calculation node Agent module and a station monitoring module for monitoring to obtain the monitoring data; accessing to a KVM (K Virtual Machine) management program through Libvirt API (Application Program Interface) of the calculation node Agent module and calling the corresponding Libvirt API to browse and obtain all the monitoring data of the cloud host; monitoring the network availability of the cloud host through at least one station monitoring module, and collecting and calculating monitoring data of a cloud platform through at least an alternate manner, and then storing the monitoring data in a database; and alarming and monitoring all the monitoring data through an alarm module according to alarm setting rules which are set by users. Through the monitoring system and the monitoring method, the users can comprehensively know the available state of the cloud host according to monitoring items and alarm strategies, thereby preventing the cloud host Agent from excessively occupying virtual resources of the cloud host when the monitoring projects are excessive.

The invention provides an instant positioning and map construction system and method with semantic perception, and the method comprises the steps that an image collection and preprocessing module obtains an RGB image collected by a camera, and carries out the preprocessing of the RGB image, and obtains a preprocessed image; a semantic segmentation module performs two-dimensional semantic segmentation on the preprocessed image to obtain a two-dimensional semantic segmentation image; the depth estimation module acquires depth information of the preprocessed image to obtain a depth prediction image of the preprocessed image; a camera pose estimation module performs camera pose estimation according to the input preprocessing image and the depth prediction image thereof to obtain a camera posematrix; and the three-dimensional dense point cloud reconstruction module carries out dense point cloud reconstruction of a three-dimensional scene on the current frame preprocessing image, the depthprediction image of the current frame preprocessing image and the current frame camera pose matrix, and maps the two-dimensional semantic segmentation image to dense point cloud of the three-dimensional scene to complete reconstruction of the three-dimensional scene with semantic information. The method has accuracy and robustness, and understanding and reconstruction of the scene are realized only by means of the monocular camera.

The invention relates to an emotion speech recognition method based on natural language comprehension. The technical scheme comprises firstly collecting emotion speech signals through a microphone, conducting pretreatment and feature extraction on the collected emotion speech signals, calculating distortion degree of the collected emotion speech signals through sentence structure emotional analysis, and determining an emotion feature vector; then respectively entering a training phase and a recognition phase; and finally conducting model matching between the emotion feature vector after the natural language comprehension and the emotion feature vector in an emotion speech base to obtain recognition results. The training phase firstly adopts emotion classification to classify the emotion feature vector, then conducts model train and finally builds the emotion speech base; and the recognition phase adopts semantic analysis and pragmatics analysis to conduct natural language comprehension on the emotion feature vector. The emotion speech recognition method is strong in naturalness of the emotion speech recognition, can integrally comprehend the emotion speech signals, and improves recognition rate of the emotion speech signals.

The invention provides a service intelligence navigation method, including the steps of: carrying out grammar matching based on metadata for word segmentation result of query proposed by a user; finding out a matched word model mode according to the word segmentation result and acquiring the service classification to which the matched word model mode belongs; carrying out matching search to the word segmentation result and descriptive information or business information and obtaining a first candidate service classification list; acquiring service from the only candidate service classification of the first candidate service classification list and returning the service to the user; if the candidate service classifications are two or more than two, carrying out fuzzy reading of non-login string in the query and obtaining a second candidate service classification list; carrying out fusion operation according to the first candidate service classification list and the second candidate service classification list and obtaining common candidate service classification; if the common candidate service classification is only one, returning the service to the user; if the common candidate service classification does not exist or is more than one, then showing the navigation is a failure.

A method for identifying Chinese full name according to Chinese short form of entity includes screening out candidate primitive set from normally used entry-bank according to abbreviation sentence to be identified, utilizing multi-path priority function combination to screen said set, calculate priority of candidate primitive, holding candidate primitive with high priority and seeking out one candidate primitive with highest priority as final result. The system used for realizing said method is also disclosed.

The invention discloses a digital body-building system which comprises a video acquisition module, a motion state acquisition module, a vital sign acquisition module, a data processing module, an action standard judgment module, an action guidance advice output module, an action command recognition module, a central processing unit, a three-dimensional teaching module, a man-machine operation module, a body-building result statistics module, a physical model building module, a virtual actuator, a virtual sensor and a simulation analysis module. The digital body-building system disclosed by the invention realizes automatic determination and guidance of body-building actions; by using the delivery function of a three-dimensional teaching video of the system, the repeated playback and learning of the teaching video or a guidance video can be performed, so that repeated viewing and understanding of body builders for each action are greatly facilitated, the body-building efficiency is improved, the actions are more standard, and the workload of teachers is also reduced; meanwhile, the comprehensive analysis on human body situations and the targeted body-building program planning can also be realized in a manner of combining a physical model and simulated analysis according to the situations of each body builder.

Provided is a satellite orbit gesture visual three-dimensional displaying method. The satellite orbit gesture visual three-dimensional displaying method comprises the steps: (1) building a three-dimensional model, reading a 3DS model in a database, checking whether models are complete or not, sending the 3DS model to a state display if the models are complete, and if the models are not complete, building relevant models, and sending a data request data package to a server and receiving the data request data package at the same time; (2) checking and comparing data in the data package, finding out data needed by the state display, analyzing the data, and dividing the data into two parts, wherein one part of data are ground equipment data, and the other part of data are on-star remote sensing data; (3) for the ground equipment data, extracting time information, calculating sun and moon ephemeris, and calculating the positions of the sun, the earth and the moon, satellite orbit data and the like to be sent to the state display; (4) for the on-star remote sensing data, classifying the data into three types to be displayed, wherein one type of data are health state data, one type of data are component power source state data, and another type of data are gesture information. The state display adopts the OpenGL to carry out three-dimensional model drawing and display.

The invention relates to a system for ensuring a computer to understand natural languages, comprising a database, a related command set, a string creating unit, a subject-predicate recognizing unit, a key word recognizing unit and a logic processing unit, wherein the database is used for storing the categories of language material resources and application programs; the related command set is used for logical operation and program control of input natural languages; the string creating unit is used for generating input strings according to the input natural languages; the subject-predicate recognizing unit is used for recognizing the subject and predicate character fields in the input strings and respectively acquiring the subject and predicate object code segments matched with the subject and predicate character fields by retrieving the database; the key word recognizing unit is used for recognizing the key words / phrases in the input strings and acquiring corresponding logical operation or program control commands from the command set according to the key words / phrases; and the logic processing unit is used for combining the subject and predicate object code segments into an integral program file. The invention also provides a method for ensuring the computer to understand natural languages. The method realizes integral recognition and understanding of the natural languages by recognizing the subjects and predicates and the logical connectives thereof in the natural languages.

The invention relates to a system for acute coronary syndrome case information collection. The system comprises a data collection side, a server side, a control side, a user side, a doctor inquiry side and a data cloud platform. The system is in butt joint with data information of a hospital information system, a clinical data center, a data warehouse and the user side and can comprehensively collect related information of the acute coronary syndrome; the doctor inquiry side is provided with identity authentication equipment, the examination information of a patient can be obtained in real time, information sharing of departments is achieved, and the diagnosis efficiency is improved. Information of doctors is opened to users, and convenience is brought to the users for independently choosing doctors and hospitals for seeking medical advices.

The present invention provides a characteristic lexical item recognizing method for a Chinese named entity, and the method includes: executing a segmentation operation on the named entity to be recognized, then getting a candidate named entity; executing an initial operation on the candidate named entity, and getting a first result; establishing a dictionary basing on the candidate named entity and the first result, and establishing a word situation dictionary basing on the first result; the dictionary and the word situation dictionary are jointly called as a dictionary database; referring to the dictionary database, executing a compounding processing on the first result time after time; after one time of the compounding processing, expanding the dictionary database basing on the processing result; the extended dictionary database is used as the dictionary to be referenced on the next compounding processing; acquiring the recognized characteristic lexical item basing on the results after a plurality of times of compounding processing. The present invention also provides a characteristic lexical item recognizing system for the Chinese named entity. Dispensing with the context, the present invention can realize recognizing and understanding on the characteristic lexical item of the Chinese named entity, and can improve the accuracy rate of the natural language understanding and the information searching.

The invention relates to an understanding system of human behaviors in intelligent space. The understanding system comprises an information perception module, an information processing module, an information scheduling module, an inference module, an information storage module and an executing module. The information perception module, the information processing module, the information scheduling module, the inference module and the information storage module are connected in sequence. The understanding system firstly collects and processes contextual information provided by various sensors in the intelligent space, redundancy and uncertainty of information provided by the intelligent space lead to uncertainty of knowledge representation, and the understanding system conducts research on uncertain knowledge representation and rule conclusion reliability. On the basis, processed multi-source information is inferred through a kernel CLIPS inference machine in a CLIPS expert system, therefore, understanding of the human behaviors is achieved, and a robot is directed to provide corresponding services for a service object.

The invention discloses a method and a terminal for displaying progress information and belongs to the technical field of computers. The method comprises the steps of: scanning progresses operated currently in a memory; classifying or sequencing the scanned progresses based on a preset progress display mode; displaying category of each progress after the classification and the progresses in each progress category, and displaying each progress category and relevant information of the progresses in each progress category; or displaying the sequenced progresses and displaying relevant information corresponding to the sequenced progresses. Through a preset progress classification mode, the progresses are displayed according to the category when a user views the progress information, so that understanding of the user for the progresses operated currently by the terminal is realized, management efficiency of the user for the progresses is improved, and then safety of the system is improved.

The invention relates to a well-drilling early overflow loss ground monitoring device. The well-drilling early overflow loss ground monitoring device comprises a data acquiring device, a data processor and a drill floor displayer connected with the data processor, wherein the data acquiring device is respectively connected with inlet and outlet flow sensors, a bit pressure sensor, a drill disc rotating speed sensor, a drill disc torque sensor, a mud pit liquid level sensor, a sleeve pressure sensor and a vertical pipe pressure sensor and is connected with the drill floor displayer through the data processor. The well-drilling early overflow loss ground monitoring device adopts a structure mode recognition method, automatically adjusts sensitivity, recognizes the type of well kick media, adopts pictorial chart outputting, achieves intelligentized monitoring, is high in monitoring accuracy, greatly shortens time for finding well kick, facilitates timely adoption of a well control measure, avoids safety accidents, is suitable for various well drilling working conditions, makes up the limitations of traditional monitoring methods, is simple in system structure, simple and convenient to maintain, low in monitoring cost and wide in application range and facilitates large-scale popularization.

The invention provides a correlation method of a 3D virtual world and a real city. The method comprises establishing a 3D virtual city by simulating the real city, and enabling members, who have physical stores in the real city, to apply for creating 3D virtual stores and to establish the 3D virtual stores in positions, which correspond to the physical stores, inside the 3D virtual city to become platform merchant; storing the relevant information of the 3D virtual stores and correlations of the physical stores into a server side. Therefore, consumers can do virtual shopping inside the 3D virtual city through a 3D virtual software platform, achieve offline consumption or ordering through the physical stores or enter the open 3D virtual stores for experience, and after registering as members, achieve online direct consumption or online ordering and offline consumption. The correlation method of the 3D virtual world and the real city can achieve correlation between a virtual world and the real world and meanwhile achieve correlation between human activities in the real world and virtual activities in the 3D virtual world.

The embodiment of the invention discloses a model display method and a device in a three-dimensional house model, a computer readable storage medium and electronic equipment. The method is applied tothe electronic equipment and comprises the following steps: determining a user visual angle range in a three-dimensional house model; according to the plane house type graph, obtaining first inter-function feature information of a local model corresponding to the user visual angle range in the three-dimensional house model; wherein the planar house type drawing and the three-dimensional house model correspond to the same house; and displaying the local model, and displaying the first inter-function feature information corresponding to the local model. Compared with the prior art, the method ofthe invention can better meet the requirement of the user for understanding the house.

The invention provides a navigation device and a navigation system for a scenic spot. The device comprises a motion sensing unit, a data processing unit, a position information acquiring unit and a display unit, wherein the motion sensing unit is used for acquiring motion sensing data of a first object in the scenic spot; the motion sensing data comprises acceleration sensing data and / or gyroscope sensing data; the data processing unit is used for calculating the motion sensing data so as to acquire a space coordinate value and a direction value of the first object; the position information acquiring unit is used for calculating to acquire position information, in a preset scenic spot navigation map, of the first object according to the space coordinate value and the direction value as well as the scenic spot navigation map; and the display unit is used for displaying a position, in the scenic spot navigation map, of the first object. By adoption of the invention, the position of a visitor in the scenic spot can be acquired and displayed to the visitor in real time so that the visitor can learn about own position in the scenic spot and plan a navigation route to realize a navigation function; and a user experience is enhanced.

The control invention discloses a speech method. The speech control method comprises following steps of an intelligent television: receiving text information sent by a voice input device, wherein text information is obtained in such a manner that the voice input device is used for speech recognition of speech signals inputted by a user, making semantic analyses of text information in accordance with a pre-set comparison table of speech standard codes in order to obtain speech information of text information, and controlling the intelligent television according to semantic information. The invention further discloses the intelligent television capable of effectively recognizing semantics of speech signals inputted by the user to understand natural semantics of speech signals and increase recognition accuracy. Therefore, the intelligent television can be well interacted with a user and user experience is improved.

The invention provides a method of intelligent remote-sensing image understanding of describing ground-object space relationship semantics. The method includes: constructing a remote-sensing image semantics understanding reference library, and carrying out intelligent semantics description of a remote-sensing image. The step of constructing the remote-sensing image semantics understanding reference library includes acquiring contents such as four parts of remote-sensing image blocks, vector data blocks corresponding to the remote-sensing image blocks, description sentences corresponding to theremote-sensing image blocks, and targets in remote-sensing images and relationship graphs among all the targets. The step of carrying out intelligent semantics description of the remote-sensing imageincludes: training a model, testing the model, and carrying out intelligent remote-sensing image description, which is based on ground-object space relationships, on a remote-sensing image block outside a database. Effects of applying the technical solution of the invention are that: understanding on the remote-sensing image is realized from different angles; the established remote-sensing imagehigh-level semantics reference library allows later further addition of description sentences corresponding to remote-sensing images, and improvement of ability of image understanding and descriptionof a computer is facilitated; and a processing method is streamlined, and realization is easy.

The present invention provides a method, an apparatus, a terminal and a sever for displaying a webpage comment. The method comprises: sending a webpage browsing request to the server; receiving webpage information corresponding to a webpage address returned from the server, wherein the webpage information comprises a source code and a comment object information list, and each piece of comment object information in the comment object information list comprises location information of a webpage passage element commented by a user; and displaying a webpage passage corresponding to a webpage address according to the source code comprised by the webpage information, and rendering and displaying a webpage comment corresponding to the webpage address according to the comment object information list. In the method provided by the present invention, the comment object information in the comment object information list comprises the location information of the webpage passage element commented by the user; the webpage comment displayed by the method can embody which part of contents is commented by the user, so that the webpage comment can be understood in combination with the webpage passage element, thereby improving reference and intelligibility of the webpage comment.

The invention discloses a lubricating oil detection device. The device comprises a detection circuit, a detection pipeline and sensors which are arranged on the detection pipeline and are coupled with the detection circuit, wherein the sensors comprise an abrasive particle sensor, a humidity sensor and a quality sensor; lubricating oil to be detected flows through the detection pipeline; the abrasive particle sensor is configured to detect oil abrasive particles of the lubricating oil; the humidity sensor is configured to detect the humidity of the lubricating oil; the quality sensor is configured to detect the oil quality of the lubricating oil; the detection circuit is configured to respectively obtain oil abrasive particle data, oil humidity data and oil quality data from the abrasive particle sensor, the humidity sensor and the quality sensor, and output a detection result according to the obtained data. With the adoption of the technical scheme, the information of the abrasive particles, water, oil products in the lubricating oil can be comprehensively known, so that operating and maintaining of equipment are facilitated, and the cost is reduced.

The invention provides a method for constructing a medical business knowledge base, a method and a system for obtaining a medical business semantic model, and a medium. The medical business knowledgebase comprises a semantic key value pair learning network, a business type learning network and a semantic model learning network, and the method for constructing the medical business knowledge base comprises the following steps: training the semantic key value pair learning network by utilizing first training data so as to output a semantic key-value pair label set based on a phrase set; trainingthe business type learning network by utilizing second training data so as to output business type labels of the medical receipts based on the set of semantic key labels; and training the semantic model learning network by using the third training data so as to output a medical business semantic model based on the set of semantic key tags and the business type tags corresponding to the set of semantic key tags. The semantic model of the medical document can be obtained, so that further applications such as fine-grained query and medical archive management are realized.

The invention discloses a stereoscopic image display machine based on interactive teaching. The stereoscopic image display machine is formed by arranging an imaging part, a folding part, a rotating part and a lifting part in sequence from bottom to top. During working, a linear push rod motor of the stereoscopic image display machine lowers the stereoscopic image display machine to be flush with a sight line in height; afterwards, a stepping motor of the stereoscopic image display machine is controlled to make a square crank drive a connecting rod to rotate, so that a guide rail elongates, and a trapezoidal frame and an organic glass plate fixedly connected to the trapezoidal frame are opened to form an angle of 45 degrees with the horizontal plane; next, the stepping motor of the stereoscopic image display machine is controlled to make the stereoscopic image display machine rotate, meanwhile, an LCD is turned on, and at this time, a stereoscopic image of the four different surfaces of a presented object can be seen at a position in a classroom through the stereoscopic image display machine. When not used, the stereoscopic image display machine is folded and lifted to a ceiling to achieve the purpose of saving space of the classroom. Besides, the stereoscopic image display machine has the advantages of being low in cost, multifunctional, high in service performance and the like.

The invention provides a method for acquiring a 3D motion trace of an object. The method comprises the following steps: setting at least one fixed object as a position reference point, wherein, one position reference point is arranged at a motion starting point of the object, and transferring position information of the position reference point to the object as the current motion reference point of the object; measuring an orientation value, a height value above a sea level and an acceleration value of the object in motion with an orientation sensor, a height value above the sea level sensor and an acceleration sensor arranged on the object; calculating the object motion trace with the orientation value, the height value above the sea level and the acceleration value in combination with the current motion reference point of the object, and storing the object motion trace in a storage part of the object; and correcting the object motion trace according to the position information of theposition reference point to obtain the corrected object motion trace when the object passes through the position reference point. The method can help obtain the position, the orientation and the 3D motion trace of the motion object anytime, and solve the problem of being lost in road and direction.

The invention discloses a simple-type smart community commodity information cloud pushing method and system. The method comprises the following steps: merchant terminals shoot and push commodity information to a cloud server; the cloud server processes the received commodity information according to the publishing authority of the merchant terminals to form publishable information; user terminals submit pushing publishing requests to the cloud server through an integrated service management server; and the cloud server pushes the information to the user terminals in the community after receiving the publishing requests respectively. Through the cloud service technology, the commodity information, comprising picture and text information and short video information, issued by the merchants is accurately and quickly pushed to a visual intercom entrance machine, a visual intercom indoor machine and a user mobile terminal of each house on a digital visual intercom platform of a smart community comprehensive service system, so that each user can watch the commodity information through the terminals above, and the system and method provide real-time convenient and fast daily life service for all families and all residents in the community.

The invention discloses an intelligent vehicle traffic scene understanding and beyond-visual-range perception system and method. The method comprises the steps of obtaining a dynamic high-precision local map in real time, enabling the dynamic high-precision local map to comprise dynamic information and static information of a monitored road condition, and carrying out the target detection of the dynamic high-precision local map, wherein the detection target comprises vehicle, pedestrian and traffic light information, conducting perspective transformation on the classified pictures, positioningvehicles and pedestrian targets in the pictures, and acquiring position information of each vehicle, enabling the vehicles on the road sections to receive the images, and determining the positions ofthe vehicles in the high-precision local map according to the position information of the vehicles, and fusing the dynamic high-precision local map information of different positions of the same roadsection to construct a high-precision map of the whole road section. Beyond-visual-range perception of the intelligent vehicle is achieved, the driving safety is improved, and the method has practical popularization value.

The invention discloses an entity combination-based mathematical natural language processing realization method and system. The method comprises the steps of establishing a mathematical entity extraction model used for extracting basic mathematical elements which constitute the mathematical entity, packaging the basic mathematical elements into the mathematical entity, performing permutation and combination on the mathematical entity, performing traversal of the mathematical relation table in one time in all combined types in the permutation and combination, and judging whether a relation can be combined or not according to the mathematical relation table so as to obtain a candidate relation set of correct recognition relations to complete processing. By virtue of the method, the mathematical entity can be mined from the mathematical problem text to combine and understand into a mathematical relation, so as to realize understanding of the mathematical problem text.

The present invention relates to a dissipative particle dynamics (DPD) simulation method for grinding liquid particle characteristics. The specific steps of the method are as follows: (1) carrying out DPD particle analysis and calculation for a single grinding liquid particle: selecting a silicon carbide particle in grinding liquid to perform analysis, and establishing a DPD particle model of the silicon carbide particle; (2) carrying out DPD particle analysis for different numbers of the grinding liquid particles: selecting 100, 200 and 300 DPD particles to perform simulation comparison, and stablishing the DPD particle model of each particle to perform DPD simulation on the particles; and (3) making a simulation conclusion: by means of DPD analysis on the grinding liquid particles, curves of the density, pressure, temperature and the like of the single DPD particle better coincide with a theoretical value. According to the method, understanding and cognition of grinding liquid particle behaviors can be effectively implemented, grinding liquid particle characteristic research and use effects are improved, better application to mechanical processing is facilitated, and better control and use are facilitated.