140results about How to "Save manpower and time cost" patented technology

The embodiment of the invention discloses a knowledge graph completion method and device, a computer device and a storage medium. The method comprises the following steps of obtaining the knowledge information to be processed; extracting a named entity and an association relationship from the to-be-processed knowledge information according to a preset first extraction rule; generating an entity tuple according to the named entity and the association relationship; and supplementing the entity tuples to a knowledge graph. Via the above method, the obtained data source only needs a small part ofannotations; the grammar style is generated according to the labeled part of data, and the unlabeled data is subjected to information extraction through the grammar style, so that the dependency on labeling in the data source can be effectively reduced, the labor cost and the time cost for adding labeling to the data source are reduced, and the efficiency and the accuracy of knowledge graph completion are improved.

The invention discloses a text-characteristic-based method for recovering documents crushed by a paper shredder. The method concretely comprises three steps of image digitization, image pretreatment and image splicing, wherein the image splicing comprises the steps of carrying out opening operation with corrosion and then expansion on fragment images, searching Chinese characters inwards along the left and right edges of each document fragment subjected to the opening operation according to a template matrix, accurately recording the position of a 'horizontal stroke' on the edge of each Chinese character in the fragment images, and splicing each two adjacent fragment images by taking the total number of the matched strokes between each two fragments as a foundation according to the positions of the ' horizontal strokes' of the Chinese characters on the left edge and the right edge of the fragment images, wherein the two document fragments with the maximal total number of the matched strokes is the adjacent document fragments; and the processes are repeated on all the fragments, and the finally obtained image is the recovered Chinese document. The text-characteristic-based method can recover the Chinese documents crushed by the paper shredder, and fills a gap in the prior art.

The invention provides an epitaxy defect analyzing structure, comprising a semiconductor substrate and an epitaxy layer, wherein an epitaxy growing region is formed on the surface of one part of the semiconductor substrate and the surface of the other part of the semiconductor substrate protected by a baffling layer protection region is exposed; and the epitaxy layer is formed on the epitaxy growing region. The invention further provides a manufacturing method of the epitaxy defect analyzing structure. The invention further provides an epitaxy defect analyzing method, comprising the following steps of: carrying out defect correlation analysis comparison on the semiconductor substrate protected by the baffling layer protection region in the epitaxy defect analyzing structure and the epitaxy layer formed in the epitaxy growing region; and judging whether an epitaxy defect in the epitaxy layer comes from an epitaxy process, an epitaxy front semiconductor substrate manufacturing process or the semiconductor substrate, so as to avoid the problem that selected samples in material selecting and epitaxy machining process are completely different due to the fact that the semiconductor substrate without epitaxy and the semiconductor substrate which is not extended are not the same semiconductor substrate.

The invention relates to a Chinese character structure-based character click verification code identification and filling method, which comprises the following steps of: pre-configuring a semantic word group library, collecting semantic Chinese word groups and adding the semantic Chinese word groups for retrieval; pre-configuring a structured Chinese character library, collecting Chinese characters and adding a structure label to a single Chinese character for retrieval; collecting character click verification code pictures, detecting areas with Chinese characters in the character click verification code pictures, cutting the areas into single Chinese character pictures, and recording area coordinates of the Chinese character pictures as filling values; pre-creating a recognition model forpredicting Chinese characters; according to the identification model, performing identification prediction on each Chinese character picture to obtain each predicted Chinese character; carrying out retrieval matching on each predicted Chinese character, an input semantic class phrase library and a structured Chinese character library, and carrying out weighted comprehensive evaluation to obtain atarget Chinese character filling sequence; and according to the obtained target Chinese character filling sequence and the regional coordinates of each Chinese character picture, clicking a verification code in a verification window of the verification code and submitting the verification code.

The invention provides a discrete simulation method for a large rocket take-off drift. The method comprises the steps of (1) building a three-dimensional model of a rocket, a mobile launch platform and a ground facility on the mobile launch platform, and carrying out assembly location according to an actual launch state; (2) carrying out simulation on a time period from rocket take-off to attention, obtaining a rocket take-off altitude, maximum drift amount and roll attitude data within the time period; and (3) carrying out discrete simulation on the time period of attention and carrying out the following processing on every moment after discretization: (3.1) searching the rocket take-off altitude, the maximum drift amount and the roll attitude data corresponding to the moment from the result in the step (2), (3.2) fabricating a circumference path along a lunching axis by employing the rocket launching axis as the circle center and the maximum drift amount as the radius and combining the roll attitude data of the rocket at the moment according to the maximum drift amount data at the moment and (3.3) calculating and recording the minimum distance between the tail part of the rocket and all attention objects according to the obtained circumference path.

The invention provides an attendance statistics method and apparatus, computer equipment, and a readable storage medium. The method comprises: first attendance data of an attendance checking machine and second attendance data of an automated office system are obtained according to a preset period; a statistic analysis is carried out on the first attendance data and the second attendance data to generate abnormal attendance information; and abnormal attendance information is sent to a corresponding client. The first attendance data include clock-in punch data, clock-out punch data and the second attendance data include business trip data, leave data, vacation data, and rest data. Therefore, employees are informed of abnormal attendance information timely and thus supplement related flow data timely, thereby improving the perfect degree of the attendance report; and the attendance report is outputted quickly and conveniently, so that the manpower and time costs are lowered.

The invention relates to a patent analysis system. The patent analysis system comprises a client platform, an application software server, a database server and a file server, wherein the client platform is connected with the application software server; the application software server is connected with the database server and the file server; the application software server comprises a patent analysis module, a tool module and a management module, and is used for realizing multiple functions of the patent analysis system; and the database server comprises an information classification module and is used for establishing classification indexes for patent information by subject. The patent analysis system provided by the invention can realize effective and intelligent analysis, and has the advantages of visual result, high analysis efficiency and comprehensive functions.

The invention provides a water bloom early warning system and method based on combination of on-line monitoring and image recognition. The system comprises an on-line acquisition device and a background early warning platform. The online acquisition device is suspended in a to-be-detected water body and comprises a physicochemical index detection template and an image acquisition module, the physicochemical index detection module is used for acquiring water quality physicochemical indexes of the to-be-detected water body, and the image acquisition module is used for acquiring images in the to-be-detected water body; and the background early warning platform compares each water quality physicochemical index with a corresponding set threshold, determines the eutrophication degree of the to-be-detected water body, evaluates the water bloom outbreak risk level in combination with the color similarity between the image in the water and a reference old water image, and performs water bloom early warning according to the outbreak risk level. Through automatic on-line monitoring of water quality physicochemical indexes and combination with an image recognition technology, while accurate water bloom prediction is achieved, the labor and time cost is reduced, and the method can be widely applied to water source monitoring places such as rivers, lakes and reservoirs and has high actual use value.

Embodiments of the invention disclose an online payment method and a system thereof. The method comprises the following steps of receiving a transaction request sent by at least one application server, wherein the transaction request carries transaction information, an application server identification and a third party server identification; according to the application server identification and the third party server identification, identifying a corresponding target application server and a target third party server respectively, and acquiring interface control information of the corresponding target application server and interface control information of the target third party server; according to the interface control information of the target third party server, sending the transaction information to the target third party server, and receiving a transaction information feedback result returned by the target third party server; and according to the interface control information of the target application server, sending the transaction information feedback result to the target application server so as to carry out transaction processing. In the embodiments of the invention, a utilization rate and payment efficiency of the servers are increased and safety of a safety system is guaranteed.

The invention discloses an Internet-of-Things device identification method and system, and a storage medium, and the method comprises the steps: extracting the types, brands and device models of various types of Internet-of-Things devices according to the crawling brand website information of the various types of Internet-of-Things devices, and forming a first Internet-of-Things device informationbase; crawling an e-commerce platform website or a supplier platform website, and obtaining the types, brands and device models of various types of Internet-of-Things devices to form a second Internet-of-Things device information base; supplementing and correcting the first internet-of-things device information base by utilizing the second Internet-of-Things device information base to form a device information base of which the device model corresponds to the corresponding brand and type; and obtaining the model of the to-be-identified Internet-of-Things device through a protocol, judging whether data information of the corresponding model exists in the device information base, and if yes, determining that identification succeeds.

The invention relates to an automatic design frame system and method for an air conditioning system. A building air conditioning system and all equipment information are digitized to achieve the wholedesign process of the air conditioning system, and the automatic design frame system comprises a model lightweight module, a BIM and building energy-consuming model (BEM) conversion module, an automatic partitioning module, a system selection module, an automatic model selection module, a system diagram generation module and a plane pipe distribution module. The air conditioning system is automatically designed based on the BIM, automation of the whole process from load calculation to equipment model selection to drawing in the design of the air conditioning system is achieved, and the laborcost of the design of the air conditioning system is greatly reduced. In actual engineering, if the building design is changed and the BIM model is changed, the air conditioning system design can be automatically completed, and repeated procedures in the design process are avoided.

The present invention relates to a patent information management system. The patent information management system comprises a client platform, an information management server, a database server and a file server, the information management server being connected with the client platform, the database and the file server; the client platform, configured to control the realization of the information management server function, consists of an information editing module; the information management server, configured to realize a plurality of functions of the patent information management system, consists of an information management module, an information destruction module, a statistics module and an association module; the database server, configured to build a classified index of patent information according to topics, consists of an information classification module; and the file server, configured to store corresponding patent information, consists of an information storage module. According to the patent information management system provided by the invention, business relation, unified management, safe destruction and statistic analysis of patent information may be effectively achieved.

The invention discloses a cold cathode florescent lamp converter, a control method thereof and a control module thereof. The cold cathode florescent lamp converter comprises a driving circuit, a controller, a detector and a frequency reduction circuit. The driving circuit is used for driving a cold cathode florescent lamp, the controller generates a control signal to control the driving circuit, the frequency of the control signal is in inverse proportion to the output / input gain ratio of the driving circuit, and the duty ratio of the control signal is correspondingly reduced when the output / input gain ratio is over a preset value. The detector is used for detecting the duty ratio of the control signal; and if detecting that the duty ratio is greater than an upper limit value, the detector sends a trigger signal to the frequency reduction circuit coupled to the detector and the controller, and the frequency reduction circuit receives the trigger signal and gradually regulates and reduces the frequency of the control signal till the duty ratio is smaller than the upper limit value according to the trigger signal. Therefore, the effect of automatically and gradually regulating and reducing the frequency of the control signal is achieved.

The invention relates to a shoulder muscle bone ultrasonic structure segmentation method based on semi-supervised semantic segmentation. The method comprises the following steps: carrying out the preprocessing of a plurality of obtained tangent plane muscle bone ultrasonic images, and obtaining a strong annotation image set and a weak annotation image set after annotation; training the generationnetwork and the discrimination network by using the tangent plane muscle bone ultrasonic images in the labeled strong labeled image set, and performing reverse optimization training on the generationnetwork by using the tangent plane muscle bone ultrasonic images in the labeled weak labeled image set to obtain a total generation network; inputting the shoulder muscle bone ultrasonic image into the trained total generative network to obtain a shoulder muscle bone ultrasonic structure segmentation image. According to the intelligent identification method for the segmented muscle bone ultrasonicimage, the manpower and time cost in the medical treatment process is reduced, and the medical treatment efficiency is improved.

The invention discloses a device and a method for realizing the shielding of different working flow products. The device comprises a receiving module, a judging module, an establishing module, a coupling module and an output module, wherein the receiving module is used for receiving an application context transmitted by an application system server; the judging module is used for judging whether an invoking mode is local invoking or remote invoking according to the application context received by the receiving module; the establishing module is used for establishing a unitive interface local realizing module and a unitive interface remote realizing module; the coupling module is used for coupling the unitive interface local realizing module or the unitive interface remote realizing module and a preset working flow unitive interface and realizing the function of the working flow unitive interface by the unitive interface local realizing module or the unitive interface remote realizing module; and the output module is used for outputting the working flow unitive interface to a user. According to the device and the method for realizing the shielding of different working flow products, the aim of shielding different working flow products is realized, and loose coupling is formed between the working flow products and an application system.

The invention relates to an automatic polaroid storing device for polaroid cleaning. The automatic polaroid storing device comprises a mechanical console, a movable cart, a polaroid cleaning machine and a polaroid collecting device. The movable cart is pushed into a U-shaped area of the mechanical console for fixation after feeding. The mechanical console adsorbs, counts and transfers polaroids to the polaroid cleaning machine for cleaning. The polaroid collecting device screens qualified products and unqualified products after cleaning. The qualified products are stacked on an upper layer and the unqualified products are stacked on a lower layer. According to the automatic polaroid storing device for polaroid cleaning, the operation process is convenient and fast, only one person is required to feed materials and collect the polaroids, only one person is required for feeding and the other person is required for polaroid collection; and manpower is saved, time cost is reduced, working efficiency is improved, the quality of obtained polaroid products is guaranteed, and the degree of satisfaction of customers is increased.

The invention relates to a novel reagent testing card of clenobuterol hydrochloride. The novel reagent testing card of clenobuterol hydrochloride comprises a reagent strip and a housing with a containing space, wherein the reagent strip is accepted in the housing. The reagent strip includes a protective film, a baseboard, and includes a sample absorbing pad, a colloidal gold pad, a nitrocellulose membrane, and a water absorbing pad, which are connected in sequence and arranged on the baseboard. The protective film is arranged on one side of the nitrocellulose membrane. The colloidal gold pad coats a clenobuterol hydrochloride antibody. The nitrocellulose membrane is provided with a test line and a quality control line, wherein the test line coats a clenobuterol hydrochloride protein conjugate and is arranged near the colloidal gold pad. The quality control pad coats sheep or rabbit anti-mouse antibody and is near the water absorbing pad. The novel reagent testing card of clenobuterol hydrochloride effectively prevents air environmental factors against affecting color development of the test line, and has advantages of good test effect and quantitative test application.

The invention discloses a grain particle counting device. The grain particle counting device comprises a mechanical device and a control module; the mechanical device comprises a guide rail, a mechanical arm, a longitudinal pulley, an infrared camera, a transverse pulley, a stirring rod, a camera support connected with the stirring rod, a motor, a motion mechanism and a support; the guide rail isinstalled on the support; the motion mechanism is connected with the guide rail through the longitudinal pulley and the transverse pulley which are installed on the side face of the motion mechanism,and the initial position of the motion mechanism is adjacent to the position of the mechanical arm; the stirring rod is installed at the upper end of the motion mechanism and moves along the guide rail with the motion mechanism; the motor is installed below the motion mechanism; the control module comprises an image processing module and a motion control module. Compared with the prior art, an artificial intelligent algorithm is applied to counting grain particles, grain is automatically counted, then the grain particle counting speed is increased, and the manpower and time cost is lowered.

The invention belongs to the technical field of molecular biology and relates to a method for rapid identification of a proliferation phenotype in a cell line after esophagus cancer functional gene knockout. By a stream type cell sorting instrument, 200 positive cell populations with carrier fluorescent tags are sorted into two holes of a 96-well plate, then FA (fragment analysis) is performed after 0h culture, and then effective editing clone of a target locus can be detected after CRISPR-Cas9 targeting of the functional gene; after culturing for 7 days, fragment analysis is performed again,a fragment analysis peak can be varied according to different cell proliferation characteristics, cells high in proliferation speed are remarkable in peaking while cell populations low in proliferation speed are not obvious in peaking or even have no peak, and thus, the proliferation phenotype after esophagus cancer functional gene knockout can be quickly judged.

The invention discloses an automatic deployment method and device for the train control center test environment, relates to the technical field of railway control system maintenance, and aims to realize automation of a test environment deployment process before a test by utilizing an automatic deployment tool on a client side, save manpower and time cost and improve deployment efficiency. The method comprises steps of obtaining a test object, configuration data corresponding to the test object and peripheral software required for testing from a data server; replacing the data information written into a lower computer chip and restarting a lower computer by utilizing the test object and the configuration data corresponding to the test object; deploying peripheral software required by the test to a test server and starting the peripheral software required by the test. The method is mainly applied to the automatic test environment deployment process.

The invention provides a cross-network-segment server OS deployment system and method. The system comprises a central node, intermediate nodes and child nodes. Each intermediate node comprises a firstnetwork card and a second network card; the central node is connected with the first network card of the intermediate node; the child nodes are connected with a second network card of the intermediate node; the first network card and the second network card are in different network segments. According to the method, complete copying of the operating system deployment time point data is performedby using a physical copying method for subsequent system reset, permanent use of one-time installation is realized, and the labor and time cost brought by repeated installation requirements is reduced.

The invention relates to the technical field of artificial intelligence. The invention discloses an order intelligent auditing method, which comprises the steps of receiving a to-be-audited order set,identifying an order type of the to-be-audited order set, arranging and summarizing the to-be-audited order set according to the order type to obtain an order report set, and extracting an auditing dimension index set of the order report set; and matching the audit dimension index set with a pre-constructed risk index set to obtain a matching result, and generating an audit result of the to-be-audited order set according to the matching result. The invention further provides an order intelligent auditing device, electronic equipment and a computer readable storage medium. According to the invention, automatic order auditing can be realized. In addition, the invention also relates to a blockchain technology, and the to-be-audited order set can be stored in a blockchain.

The invention discloses a remote sensing image change detection method. The method comprises the following steps: acquiring a remote sensing image; inputting the remote sensing image into a pre-generated deep learning model, wherein the depth model comprises an irregular image object depth feature extraction module and a depth feature fusion classification module, the depth feature extraction module is generated by pre-training an unsupervised stack type noise reduction automatic encoder, and the depth feature fusion classification module is composed of a pre-trained noise reduction automaticencoder, a cascade layer, a full connection layer and a logistic regression layer; receiving a change detection result graph generated by the deep learning model; and outputting a detection result according to the transformation detection result graph. In the training process of the learning model, a large amount of labeled data does not need to be used for training, so that the labor and time cost of deep learning is reduced, the edge and shape information of the irregular object can be kept by the deep learning model, and the depth features of the irregular image object can be expressed anddescribed.

The invention discloses an online telemetering method of the reading of a mechanical SF6 gas density meter. The online telemetering method includes the image acquisition step, the image reading information recognition step and the reading step. Accordingly, the invention further discloses an online telemetering device of the reading of the mechanical SF6 gas density meter. The online telemetering device comprises an image acquisition unit, an image analyzing and processing unit and a reading end device, wherein the image acquisition unit acquires images of a dial of the mechanical SF6 gas density meter according to remote acquisition instructions, and the acquired images of the dial are digitalized into image signals to be output; the image analyzing and processing unit is connected with the image acquisition unit, receives and recognizes the image signals output by the image acquisition unit and extracts image reading information contained in the image signals; the reading end device is in remote data connection with the image analyzing and processing unit so as to receive the image signals and / or the image reading information transmitted by the image analyzing and processing unit. Through the method and device, the reading of the mechanical SF6 gas density meter can be telemetered online, and thus labor and time cost are saved.

The invention discloses an advertisement insertion time point detection method and device, and the method comprises the steps: intercepting a target video segment corresponding to a video time period where an advertisement can be inserted from a target video according to the type of the video, and dividing the target video segment into a plurality of sub-video segments through employing a sub-video algorithm, determining all candidate key frame groups on the basis of the edge key frames of each sub-lens video clip, and screening out candidate key groups of which the sound event tags of the sound clips are mute and the caption tags of the caption clips are caption-free from all the candidate key frame groups as target key frame groups; and based on the key frame group weight value of each target key frame group and the similarity score representing the video similarity measurement standard, determining the position of the target key frame group with the highest sorting value as the advertisement insertion time point as the final advertisement insertion time point. According to the invention, the advertisement insertion time point in the movie and television program is automatically detected, so that the labor and time cost is reduced.

The invention discloses a data scene generation method and device and computer equipment, and the method comprises the steps: obtaining target data which comprises a type field; determining a plurality of data interfaces corresponding to the target data according to the type field; determining a target data interface corresponding to each data interface according to the plurality of data interfaces corresponding to the target data and a target logic algorithm; determining a link relationship among the target data interfaces according to a target algorithm; and generating a target data scene according to the target data, each target data interface and the link relationship. By implementing the method and the device, automatic matching of the current target data interface and the next data interface, automatic series connection among the target data interfaces and generation of the target data scene are realized, manual data construction is not needed, the labor time cost is saved, and the generation efficiency of the target data scene is improved.

The invention discloses an express transceiver system based on an unmanned aerial vehicle. The express transceiver system consists of a mobile user terminal, an express operation control terminal and an unmanned aerial vehicle system, wherein the express operation control terminal comprises a data transceiver, a transmitting code generation module, a receiving code generation module, a controller and a wireless communication module I, the mobile user terminal is connected with the data transceiver, uplinks of the transmitting code generation module and the receiving code generation module are connected with the data transceiver, downlinks are connected with the controller, the controller is connected with the wireless communication module I, the unmanned aerial vehicle system is connected with the express operation control terminal through wireless data, and the unmanned aerial vehicle system consists of a flight control system, a navigation module, a route generation module, a wireless communication module II, a miniprinter and an input module. With the adoption of the express transceiver system, the express transmitting and receiving efficiency is improved, the labor and time cost is reduced, and the safety and privacy are high.

The invention relates to an information processing method for automatic design of a centralized air conditioning system, comprising the following steps: step S1, loading a building model file, and calculating a load curve based on the loaded building model file; Step S2: Generate the pipe network of the air-conditioning system and the equipment selection requirements of the air-conditioning systembased on the building model file and the calculated load curve. Compared with the prior art, the invention is directly designed according to the architectural BIM model, is closely connected with thearchitectural design link, reduces the communication and interaction cost between different specialties, and if the architectural design is changed, the HVAC design can be changed correspondingly immediately.

The invention relates to the field of medical images, and discloses a system and a method for DICOM-based (Digital Imaging and Communications in Medicine) on-demand printing. The system comprises a server, patient information acquisition equipment, an ink-jet printer, a film and ink. During output of medical images, a patient information carrier is read through the patient information acquisition equipment; the patient information is matched through network communication and the server; identification information and inspection information of the patient are found to call out corresponding medical images; the medical image is sent to the printer for printing to realize on-demand printing output. Adopting the medical image printing system provided by the invention can realize on-demand printing in a controllable manner and make the DICOM-based medical image output more simplified and more reasonable, so that manpower and time costs are saved.