83results about How to "Solve access difficulties" patented technology

The invention discloses a patient similarity metric transfer system among disease domains on the basis of transfer learning, and belongs to the technical field of the artificial intelligence software of computers. By use of the system, four submodules including a data preprocessing submodule, a similarity metric evaluation submodule, a similarity metric learning submodule and a similarity metric transfer submodule are constructed to finish similarity metric in a disease domain and patient similarity transfer among disease domains. The method aims to overcome the problem that traditional metric learning can not effectively work under a situation that supervisory information is difficult in acquisition and a patient sample amount in a specific disease field is small. The system can provide service support for a current accurate medical scene.

The invention relates to a modeling method of a plane secondary envelope torus worm-drive worm gear hob tooth profile. The method comprises the following steps: 1) giving initial design parameters; 2) calculating sum which corresponds to an initiating terminal; 3) obtaining point coordinates on an A tooth face; 4) judging whether the point coordinates are located in the working range of a hob or not, if yes, turning to the step 5, and if not, turning to the step 8; 5) obtaining the point coordinates on the A tooth face; 6) obtaining a coordinate of a middle point; 7) modifying a value, turning to the step 3, and obtaining intersecting lines of a series of intersecting planes and a tooth face of a worm gear hob basic worm; 8) modifying a tail end, and obtaining coordinates of points under the tail end and a middle point; 9) obtaining coordinates of points on a B tooth face and a middle point, wherein the B tooth face is the same with the A tooth face in the tail end; 10) obtaining point coordinates on a tooth root circular arc face; 11) obtaining point coordinates on a main cutting edge; 12) building an axial tooth profile model of a worm gear hob; and 13) building a curved face model of the worm gear hob basic worm. The method is free of building of a solid model, easy to achieve, low in modeling cost, and high in modeling precision.

The invention relates to a sound-groove identification method based on emotion compensation. The emotion compensation includes three portions of emotion detection, character compensation and emotion expansion, comprising of calculating voice emotion factors to be according to the emotion detecting technique, compensating the voice change caused by emotion change respectively from the two layers of character and mode and finally improving robustness of the sound-groove identification technique to the emotion change. The invention has the advantages that the invention breaks through the inconsideration of sound-groove emotion change of the existing sound-groove identification technique, deals with the voice change caused by emotion change from the two layers of character and mode and strengthens resisting power to the voice emotion drift. The character layer standardizes the voice feature within the modeling ability of the training model by means of emotion degradation, normalization and barrier to reach the purpose of inhibiting the influence of the user emotion on the identification property. The mode layer obtains large scale emotion voices by employing the reverse way of synthesizing emotion voice by emotion changing rule, thereby greatly expanding the modeling information of the sound-groove model and resoling the difficulty of obtaining emotion data.

The invention relates to the technical field of computer application, in particular to an enhanced generated countermeasure network and a target sample identification method. At least one enhance generator and at least one enhanced discriminator are provide for in that enhanced generated countermeasure network of the present invention, the enhanced generator processes the obtained initial data togenerate data, the enhanced discriminator comprises a rolled base layer, a base capsule layer, a convolution capsule layer and a classification capsule layer, wherein the rolled base layer, the base capsule layer, the convolution capsule layer and the classification capsule layer are sequentially connected.

The invention relates to a GNSS and WIFI integrated positioning method and a system thereof. According to the method, through combination of at least three beacon nodes located via GNSS location-based service or at least three reference nodes located via WIFI, the relative position information of the current node is calculated, and then the relative position information is uploaded to update the position information with a timestamp in a central server and is broadcasted so as to provide convenience for other nodes to search and use. Mobile equipment indirectly obtains self position information via an RSSI locating mode after receiving the surrounding node position broadcast messages so as to achieve integrated positioning. The integrated positioning mode solves the problem that the position of the WIFI hot spot itself is hard to obtain and has great effects in the aspect of improving positioning precision and strengthening practical usability.

The invention discloses an emotion analysis method and device, and relates to the technical field of processing of natural languages. By means of the emotion analysis method and device disclosed by the invention, the emotion analysis accuracy rate is increased. The technical scheme in the invention comprises the following steps of: extracting an emotional word sequence from a target text, wherein the emotional word sequence includes sequentially extracted emotional words and non-sememic words; generating an emotional word sequence vector corresponding to the emotional word sequence; according to a pre-set classification model, obtaining an emotional mark corresponding to the emotional word sequence vector of the target text, wherein the corresponding relationship between the emotional word sequence vector and the emotional mark is stored in the pre-set classification model; and taking the obtained emotional mark as an emotion analysis result of the target text. The emotion analysis method and device disclosed by the invention are mainly used for analyzing emotion of the target text.

The invention discloses a left-atrial-appendage CT image segmentation method, and relates to medical image processing. Firstly, left-atrial-appendage CT images are denoised, prior knowledge is utilized to construct missing boundary information of a left atrial appendage, and a region growing algorithm is used to obtain a single initial segmentation image to use the same as a single Atlas label graph; then information relationships of the single label graph and an adjacent left-atrial-appendage CT sequence are utilized for improved Atlas segmentation to obtain an initial segmentation sequence;contour information in the initial segmentation sequence is used as initial contours of a level set algorithm, and approximate Euclidean distance information of left-atrial-appendage CT sequence graphs is obtained to be used as feature input; and finally, a precise segmentation result sequence is obtained through a fast level set algorithm to achieve the purpose of left-atrial-appendage CT image segmentation.

The invention relates to a film playing information feedback system and method for a second-level film theater chain. The playing information feedback system comprises theater party servers, playing control devices and a film source server. The theater party servers are arranged at theater parties and comprise film playing information databases. The playing control devices are arranged at the theater parties, are connected with the theater party servers and transmit film playing information to the film playing information databases of the theater party servers when users operate the playing control devices to carry out video-on-demand on the films. The film playing information at least comprises playing device information, film title information, playing start time and playing end time. The film source server is in network connection with the theater party servers and receives the film playing information sent by the theater party servers. According to the method, the problem that issuers are difficult to obtain the film playing information is solved.

The invention discloses a method for automatically leading and gumming shoe uppers in the technical field of shoemaking. The method specifically comprises the following steps of S1, drawing a route line on a sample shoe upper surface in advance, and obtaining point cloud data through an image acquiring device mounted on a manipulator; S2, performing splicing fusion treatment on the point cloud data, and separating route point cloud data in the route line; and S3, transforming position information and space normal vector information in the route point cloud data into a manipulator xyzrpy coordination group to lead the manipulator to perform gumming. Through the adoption of the gumming method disclosed by the invention, the gumming route can be accurately identified, repeated attempting andverifying are not needed, the operation is simple and high in efficiency, requirements for operators are low, and besides, the gumming route is high in accuracy and high in consonance.

The invention discloses a design method of a TDICMOS imaging unit, relates to a design method of a TDICMOS imaging unit, solves the problems that the existing TDICMOS imaging unit is complex in designand poor in stability. A controller unit and a detector unit are included, the controller unit generates a control signal required by the detector unit and receives the digital image data output by the detector unit; the detector unit comprises a TDICMOS detector, an operational amplifier, an LDO, a driver with a Schmidt function and a level shifter with a Schmidt function. According to the invention, power supply of a pixel part with pull-fill current, transmission of a digital input control signal, a generation mode of a charge transfer driving signal, a placement mode of a via hole in a high-density region and processing of a multilayer grounding plane are designed respectively; the utilization area of the circuit board is increased to the maximum extent, the integrity of control driving signals and a power supply is improved, and meanwhile voltage breakdown, pseudo soldering and other risks are avoided.

A movable pressure collecting ball based on bluetooth data transmission comprises a counterweight bin, an NFC label, bluetooth, an explosion-proof battery, a single chip microcomputer, a pressure sensor and an error correcting temperature sensor (selectable). The collecting ball can flow along with media in a pipeline and records the data values of time and pressure at set intervals, after the collecting ball is taken out from the tail end of the pipeline, the bluetooth is used for transmitting data to a computer, and the pressure values of the pipeline at different positions are acquired through calculation of the data such as the length of the pipeline and the time. The movable pressure collecting ball can effectively solve the problem that the whole line pressure data of most of pipe sections of existing pipelines which are deeply buried, erected overhead, arranged in water and the like are difficult to acquire, can achieve whole pipe section pressure variation detection under the premise that production is not influenced, and non-blind area whole line detection to most of the pipe sections of the existing pipelines can be achieved.

The invention relates to an underwater topographic data identification and service method and system based on deep learning, and the method intelligently extracts lake underwater topographic data by abuilt-in deep learning model to form an underwater topographic database, and provides data service for the outside; the constructed system can accurately retrieve the underwater topographic data of the corresponding area according to the lake name or coordinate place input by the user, and the user can download the detailed geographic data of the lake and the surrounding from the system. Meanwhile, the system supports a user to upload lake underwater topographic data which is not covered by the system on line, background personnel upload the lake underwater topographic data and indicate datacontributors after carrying out related auditing, the lake underwater topographic data are used by people, and a certain amount of rewards are given to data providers. According to the invention, theproblem that the current underwater topographic data is difficult to obtain is solved, the requirements of related scientific research institutions and individuals on the underwater topographic data can be greatly met, and the market prospect is wide.

The invention discloses an acquisition method, apparatus and system of electric energy meter information. The method includes: scanning an identification code corresponding to an electric energy meter and reading a character string included by the identification code; sending the character string to a server so that the server searches the corresponding electric energy meter information according to the character string; receiving the electric energy meter information returned by the server; and displaying the electric energy meter information. By employing the method, operation personnel can rapidly check all the electric energy meter information of the electric energy meter during replacement and installation of the smart electric energy meter, the replacement and installation efficiency of the electric energy meter is effectively improved, the possibility of wrong wiring is reduced, and the technical problem of easy errors of artificial check of the electric energy meter information due to difficult acquisition of the electric energy meter information in the prior art is solved.

The invention belongs to the field of face recognition, and particularly relates to a face recognition method based on a stacked convolution sparse denoising auto-encoder. The method comprises the following steps: alternately connecting a plurality of convolution sparse denoising auto-encoders with a plurality of pooling layers so as to construct a stack convolution sparse denoising auto-encoder model; removing the convolution form of the stack-type convolution sparse denoising encoder model, and training a stack-type sparse denoising auto-encoder model in a blocking mode; forming a convolution filter by using the trained parameters, and realizing a stacked convolution sparse denoising auto-encoder in a convolution form; performing face feature extraction by using a stack convolution sparse denoising auto-encoder, and classifying and recognizing faces by using a classifier. According to the invention, through the deep convolutional neural network which can be trained in an unsupervisedmanner, the face features having similar performance with the supervised deep convolutional neural network are extracted, and the training difficulty is effectively simplified.

The invention provides a method for extracting and purifying cordycepin from cordyceps millitaris mycoderma. The method comprises the steps that the cordyceps millitaris mycoderma is taken as raw materials and smashed through a smashing machine, screened and put in a vacuum drying box to be dried for standby use; a mixed solution of methyl alcohol and water is added in the dried cordyceps millitaris mycoderma powder for steeping; steeped suspension is arranged in a backflow device, backflow is heated and filtered, and an extraction solution is obtained; the extraction solution is extracted through petroleum ether, and a petroleum ether layer is abandoned; the lower remaining solution is decompressed and concentrated to recycle a solvent, and extract of the extraction solution is obtained; after the extract is dissolved through the methyl alcohol, elution is carried out through a silicon gel chromatography column, chloroform and the methyl alcohol are in flowing phases, eluent is recycled, and TLC tracking detection is carried out; the obtained eluent with the cordycepin is decompressed and concentrated, and then is heated through proper methyl alcohol or ethyl alcohol to be dissolved, and low-temperature crystallization is carried out. The technology is easy and controllable, the extraction time is short, the consumption of the organic solvent is little, the organic solvent can be recycled, cost is low, and the obtained cordycepin is high in purity.

The invention discloses a mobile device tour guide system based on GPS, wifi and Bluetooth. The problems that scenic spot introduction information is difficult to obtain and the information content issingle in the tour process of a user are solved. The equipment positioning is performed based on the basic functions of GPS, wifi, Bluetooth and the like of the mobile equipment, and the position information of the mobile equipment of the user is combined with a self-defined scenic spot map so as to acquire more accurate position information. The detailed introduction information of characters, pictures, voice, video and the like of each scenic spot is marked on the scenic spot map, and automatic explanation and recommended route navigation can be realized by means of the positioning and navigation functions. The problem that the scenic spot introduction information is difficult to acquire when the user travels can be solved, and the problems that a traditional tour guide system cannot position in real time and the scenic spot information content is single can also be solved. Therefore, the user experience of tourists in the tourism process can be improved.

The invention belongs to the technical field of Internet of Things, discloses a mass Internet of Things equipment access and management method, and designs a unified equipment abstract model. The equipment abstract model is used for describing the capability characteristics of the equipment, and the abstract model is used for describing the capability of the equipment and the provided service; and the equipment model is described in the forms of JSON and XML, and the model is uploaded to a cloud system and an equipment end after the equipment model is developed. And the capability characteristics of the equipment are equipment model identification attributes: equipment type, manufacturer, model, protocol type, service supported by the equipment, attribute and control command. According to the method, the steps are simple and clear, the massive heterogeneous equipment is subjected to standardization processing by designing the equipment unified description model, and a very good effect can be achieved. Compared with the existing Internet of Things access technology, the mass Internet of Things equipment access and management method provided by the invention fundamentally solves the problems of difficulty in Internet of Things equipment access, high cost, poor safety and difficulty in interaction.

The invention discloses a manufacturing method of an artificial rock core for simulating borehole drilling. The manufacturing method comprises the steps of: 1) manufacturing a borehole drilling paraffin blank; 2) manufacturing an artificial rock core blank containing mixed slurry and the borehole drilling paraffin blank prepared in the step 1), wherein the mixed slurry comprises quartz sand, cement, clay and water; and 3) heating the artificial rock core blank prepared in the step 2) to obtain the artificial rock core for simulating borehole drilling. The manufactured artificial rock core is similar to a natural rock core in the aspects of permeability, porosity and other physical properties, and simulates perforating measures in the oil production process; the artificial rock core manufactured by adopting the manufacturing method can be used for researching a damage transformation rule of oil production measures such as perforation fracturing on a stratum, and conducting related indoor simulation experiments; in addition, the whole manufacturing method is simple and safe, raw materials are economical and easy to obtain, and the problem that experimental analysis is limited due tothe fact that natural rock cores are difficult to obtain, unstable in rock physical property, large in complexity and anisotropy difference and the like is solved.

The invention provides an artifact-free data and artifact-containing data generation method and system and a storage medium, and the method comprises the steps of carrying out the small sample expansion of real artifact-free data based on a generative adversarial network model, and obtaining massive artifact-free data; and generating artifact simulation data based on the artifact generation methodand the real artifact-free data, and then generating massive artifact data by using a generative adversarial network model based on the generated artifact simulation data and the real artifact data.According to the method, the problems that medical CT artifact-free and artifact-containing samples are difficult to obtain, the data size is small, the diversity is low, a large amount of private data is involved and the labeling cost is high are solved, and artifact-free and artifact-containing data generation is beneficial for deep learning to carry out artifact recognition.

The invention provides a multi-lane non-contact axle number automatic calculation method and system. The method comprises a laser sensor setting scanning step, a laser radar data acquisition step, a single vehicle data separation step and an axle number calculation step. Specifically, based on laser radar data, the height and transverse coordinates of a traveling vehicle at each rotation angle are calculated by using the principle of diffuse reflection of the surface of the traveling vehicle irradiated to a lane, and then a vehicle three-dimensional model is constructed; single vehicle data is extracted, automatic calculation of the number of vehicle axles on multiple lanes is achieved under the non-contact condition, and the accuracy of vehicle type classification is improved.