56results about How to "Reduce attention" patented technology

This invention automates a portion of the existing process by applying computerized comparative assessment program to the image presented on the operator's viewing screen. This is accomplished by creating a computerized image database of all known weapons, dangerous articles, and other prohibited items. At the same time the scanned container's contents data is converted to the image presented on the operator's viewing screen, the same data is computer compared to the weapons database. Just as the human mentally compares learned weapons database against what he or she sees on the monitor, the computer accomplishes the same process, except much more timely and accurately. When the computer identifies data from the scanned image that correlates to it's prohibited items database, it alerts the security operator by any number of common means and highlights the image on the security operator's monitor.

The invention prevents a method for realizing video quantification. Through the steps of video slitting, counter setting, snapshot statistics, data aggregation and curve drawing, the method realizes concrete quantification inside the video, obtains the access data statistics of each clip, calculates the total broadcasting time and the CS time of each clip according to the data statistics, and draws the corresponding video quantification chart CV / CS curves. From the CS curves, the broadcast-on-demand quantity of each clip can be seen, and from the CV curves, the total broadcast-on-demand time of each clip can be seen. The invention can provide advertisers with reference and data support to make decision for delivering advertisements and to know the part which the users are interested in, and helps the media and the advertisers to deliver video advertisements more accurately, so that the advertisements can be put on the clip mostly viewed by the users, thus improving the performance price ratio of the advertisement and bringing more economic benefits for the advertisers. The video content provider and the video webmaster can analyze the attractiveness of the video content to the users, the user experiences and the user view behaviors so as to further improve the quality of the video service.

The invention relates to an X-ray weld defect detection method based on a convolutional neural network, and the method comprises the following steps that a weld image data set containing different types of weld defects is built, and all weld images in the data set are marked with weld labels; an AF-RCNN model is established, wherein the AF-RCNN model comprises a backbone network module, a region generation module and a target classification and position regression module; the backbone network module adopts a residual network (ResNet) and a feature pyramid network (FPN) structure, an efficient convolution attention module is introduced between the residual network (ResNet) and the feature pyramid network (FPN) to enhance the learning ability of the network for unobvious defects and small target features, and a CIOU loss function is introduced to enhance the positioning ability of aiming frames; and an AF-RCNN model is trained by using the established data set, wherein the AF-RCNN model is used for classifying and positioning weld defects. The accuracy of all defects reaches 94% or above, and the detection speed is 11.65 FPS.

The invention discloses a mobile electronic commerce information publishing and interaction platform, which comprises a mobile phone terminal, a server and sites, wherein the mobile phone terminal acquires product information embedded with a mobile phone number; the server receives and stores the product information, extracts keywords from the product information, expands the product information according to the keywords and product attributes in the server, compiles the product information into a data format capable of being identified by the sites according to the information publishing rules of each site, publishes the product information, receives the interaction information of the sites and simplifies and transmits the interaction information to the mobile phone terminal; and the sites are used for publishing the product information and generating the interaction information. The mobile electronic commerce information publishing and interaction platform can enable a user to finish the publishing, management and deletion operation of the product information on a plurality of sites on a mobile phone and perform product trading activities, transfers electronic commerce onto the mobile phone and effectively improves the operation convenience of the electronic commerce.

One embodiment of the invention discloses a method and device for controlling a mobile terminal and a mobile terminal. The method comprises obtaining layer content of a currently displayed layer; determining an object satisfying a set condition according to the layer content, wherein the set condition is that the attention on the object by a user is lower than a threshold and the object comprises layers, application programs or interface elements; using a set processing mode to perform power consumption reduction processing on the object, wherein the processing mode comprises at least one of reducing the drawing frame rate of the object, reducing the synthesis frame rate of the object, reducing the refresh rate of the area where the object is located and reducing the resolution of the object. By means of the method and the device of the invention, power consumption on objects having less attention from users can be reduced so as to reduce power consumption of terminals and give consideration to display effects, therefore increasing the endurance of terminals.

The invention provides a commodity leasing management device, a commodity leasing management method, and a commodity leasing management system, and relates to the commodity leasing field. The commodity leasing management device, the commodity leasing management method, and the commodity leasing management system are used to generate consumption preference information and low attention result information according to times and frequency of commodities of various types in a certain time period, and is used to generate a consumption habit analyzing report according to the consumption preference information, the low attention result information, commodity type information, and first position information, and therefore managers are aware of high attentions of customers to which consumption areas and which commodity types and the low attentions of the customers to which consumption areas and which commodity types by using the consumption habit analyzing report, and can master the preferences of the customers, and can rearrange, plan, and adjust the various consumption areas, and then the requirements of the customers are better satisfied by the various consumption areas after the rearrangement, the planning, and the adjusting.

An electronic control module interface system may include a sensing surface having a symbol representative of a function of a motor vehicle, a sensing device coupled to the sensing surface, configured to sense a proximity between a physical pointer and the sensing surface and to generate a browsing signal when the proximity is less than a predefined proximity threshold value, and configured to generate a selection signal when the physical pointer contacts or depresses the sensing surface, and an output device coupled to the sensing device, and configured to produce an output representative of the function of the motor vehicle upon receiving the browsing signal.

Provided are an advertising independent demonstration method, a system and an advertising machine and application, relates to the field of advertising machine development and advertising display, andaims at solving the problem of low attention in the advertising machine of the prior art. The method comprises a step of triggering attention and a step of summoning when an audience leaves; when theaudience enters into an identification area, a response to the attention of the audience is made, to attract the attention of the audience, and the content of the advertisement for the audience is pushed; the step of summoning when the audience leaves: when the audience leaves the identification area, a summoning response is sent, to attract the audience back or return to the identification area and continue to focus on the advertising machine. The invention further provides a virtual device corresponding to the method. An advertising machine and a computer readable storage medium are furtherprovided to realize the method. A computer program based on the method is implemented by the processor of the advertising machine host to realize the steps of the method. The method is also applied tothe sale or display of electronic products with a display screen. The attention rate of the advertising machine is greatly improved.

The invention discloses an intelligent community safety monitoring system and a safety monitoring method. The system comprises a monitoring cloud, a video monitoring unit, an entrance monitoring unit, an outlet monitoring unit, a field prompting unit, an automatic alarm unit and a hand-held terminal. The monitoring cloud is arranged in an equipment operation room; the video monitoring unit is connected with the monitoring cloud through an electric signal transmitted in the electronic connecting line; the entrance monitoring unit is installed at an entrance of a community and is connected withthe video monitoring unit via an electric signal transmitted in the electronic connecting line; the exit monitoring unit is installed at a community exit and is connected with the video monitoring unit through an electric signal transmitted in the electronic connecting line, the on-site prompting unit is installed on the community monitoring device, and the automatic alarm unit is connected withthe monitoring cloud through an electric signal transmitted in the electronic connecting line. According to the smart community safety monitoring system and the safety monitoring method, the identities of people entering and exiting the community can be identified in time, and the community comfort level index is improved.

An automated “Voice Search Message Service” provides a voice-based user interface for generating text messages from an arbitrary speech input. Specifically, the Voice Search Message Service provides a voice-search information retrieval process that evaluates user speech inputs to select one or more probabilistic matches from a database of pre-defined or user-defined text messages. These probabilistic matches are also optionally sorted in terms of relevancy. A single text message from the probabilistic matches is then selected and automatically transmitted to one or more intended recipients. Optionally, one or more of the probabilistic matches are presented to the user for confirmation or selection prior to transmission. Correction or recovery of speech recognition errors avoided since the probabilistic matches are intended to paraphrase the user speech input rather than exactly reproduce that speech, though exact matches are possible. Consequently, potential distractions to the user are significantly reduced relative to conventional speech recognition techniques.

The embodiment of the invention discloses a semantic recognition method, a semantic recognition device and a storage medium. The method comprises the steps of obtaining a dialogue statement, converting the dialogue statement into text information, dividing the text information into N words through a dictionary, and converting the N words into N corresponding word vectors, wherein the dictionary comprises words appearing in the collection dialogue; performing semantic recognition on the N word vectors by using a neural network added with a bidirectional long-short-term memory unit, and giving asemantic recognition result. By adopting the embodiment of the invention, the semantic recognition precision can be improved.

The invention provides a power consumption optimization method and device for a 5G terminal during service concurrence, electronic equipment and a storage medium, and the 5G terminal is located in EN-DC networking. The method comprises the following steps: detecting whether the 5G terminal is in a state of concurrence of a voice service and a data service; and if so, determining whether the 5G terminal satisfies a preset power consumption optimization condition, and if so, switching the data service to a 4G LTE network. According to the method, the power consumption of the 5G terminal can be remarkably reduced, and the user mainly concentrates on the voice service and the attention to the data service is reduced under the scene of the voice service and the data service, so that the data service is switched to the 4G LTE network and the user experience is not greatly influenced.

The embodiment of the invention provides a voice recognition method and device and a device for voice recognition. The method specifically comprises the steps: determining voice features of to-be-recognized voice; determining a voice recognition result corresponding to the voice feature by using an acoustic model, wherein the acoustic model comprises a neural network model and a hidden Markov model, the neural network model comprises an input layer, a hidden layer and an output layer, and the hidden layer comprises a feedforward neural network layer and a self-attention neural network layer; and outputting the voice recognition result. The voice recognition method and device and the device for voice recognition can improve the voice recognition efficiency.

The invention relates to an air bag type artery compression hemostasis control system and corollary equipment thereof. The air bag type artery compression hemostasis control system comprises a controlmodule, and the control module is connected with a sensor, a display interaction panel and an air pump used for inflating and deflating a compression hemostasis air bag; the sensor monitors the bloodpressure and oxygen saturation of a human body in real time and also monitors the pressure of an output pipeline of the air pump; and the display interaction panel comprises a display screen and keys. The system has the beneficial effects that radial and ulnar artery compression hemostasis and vital sign monitoring are integrated; the changes of blood pressure, oxyhemoglobin saturation and heartrate of a patient are monitored in real time, the compression strength of the compressor is dynamically adjusted according to the blood pressure, accurate and intelligent compression hemostasis is achieved, adverse events such as postoperative hypotension, maloperation reflex and syncope can be monitored, and various postoperative complications are effectively reduced.

The invention discloses a footprint pressure image retrieval method. The method comprises the following steps: 1, collecting and preprocessing a footprint pressure image data set; 2, establishing a feature extraction network composed of K layers of multi-scale self-attention convolution modules; 3, establishing an incomplete scoring module composed of a global feature branch and an incomplete scoring mask branch; step 4, establishing a feature comparison module composed of a common visible feature extraction module, a local feature pooling module and a triple loss function; and step 5, carrying out network training, parameter optimization and testing. The multi-scale self-attention convolution footprint pressure image retrieval method is adopted, discriminative features of footprint pressure image retrieval can be effectively extracted, and meanwhile, for the problem that footprint images are incomplete, an incomplete scoring model is adopted, so that the attention of a network to incomplete parts can be reduced, and the influence of the incomplete image on the retrieval process is effectively inhibited.

The invention discloses an attention-based joint source channel method for a text, which uses a generator comprising an encoder, a channel and a decoder to carry out joint source channel coding on the text, and uses a GRU to replace an LSTM, so that the calculation speed is higher. Meanwhile, an attention mechanism is added, so that the decoding efficiency and accuracy are improved. By using the GAN framework, the parameter update of the generated model is not directly from an original data sample but from the back propagation of a discrimination model. According to the method, a one-way text decoding framework based on GRU is used as a generation model, a CNN is used as a discrimination model, and samples generated by a generator are discriminated. Point mutual information is added in beam search, and words are decoded more accurately together with the maximum likelihood probability. At a decoding end, a synchronous bidirectional decoding method is used. Meanwhile, L2R and R2L decoding is carried out, and alignment and information interaction are realized by utilizing SBAtt and bidirectional beam search, that is, past and future information is utilized at the same time, so that the word error rate of decoding is reduced.

An instrument operation apparatus 1 includes a tag reader 5 and a manipulating device 3. According to a user's manipulation to the manipulating device, the instrument operation apparatus 1 transmits a command signal to one of instruments (101, 103, 105) to operate. The manipulating device (3) includes a detection portion (9) to detect the user's manipulation, and a device information transmission portion (11) to transmit device information according to the manipulation detected by the detection portion using a wireless communications technology. The tag reader (5) includes a reception portion (13) to receive the device information; a control circuit (15) to determine a command signal according to the device information received via the reception portion, and a command signal transmission portion (17) to transmit the command signal to the one of the instruments. The instrument operating apparatus is characterized in that an arrangement position at which the manipulating device (3) is arranged or attached is enabled to be changed.

The embodiment of the invention provides a data synchronization method, device and equipment, and a storage medium. According to the data synchronization method provided by the embodiment of the invention, when a data synchronization request initiated for a target data object is received, a data synchronization plan corresponding to the target data object can be obtained, and then, according to the data synchronization plan corresponding to the target data object, the data corresponding to the target data object is transferred from the source resource to the target resource, thereby a data synchronization process of a data object level is realized. The method is advantaged in that the data object is taken as a data synchronization unit, so the attention to the data table can be effectivelyreduced, and data synchronization efficiency can be improved.

The invention discloses an aluminum-magnesium alloy wire drawing machine wire breaking brake device and method. The device includes: a wire-releasing device, a wire drawing die, a wire-drawing device, a wire-receiving device, a base, a PLC controller, a sensor, and a walking track; The braking method is as follows: if the wire is broken and the wireless material passes through the sensor, the sensor transmits an electrical signal to the PLC controller, and the PLC controller sends a brake signal to the wire pulling device and the wire taking device, and the wire pulling device and the wire taking device stop working and send out The alarm notifies the staff, the sensor walks on the walking track, and automatically finds another breakpoint. The invention ensures the safety of the equipment, shortens the wiring time, improves the production efficiency, and reduces the attention and work intensity of the staff.

The embodiment of the invention discloses a voice interaction method and device and a storage medium, and the method comprises the steps of building a user portrait of a client, and determining the classification of the client according to the user portrait; based on the classification of the client, making a collection strategy, and carrying out voice communication with the client according to the collection strategy; converting the client voice received during the voice call into the text information, and then generating a word vector; and analyzing the word vector by using a long and shortterm memory network to obtain an analysis result of the client voice. By adopting the embodiment of the invention, the precision of the voice analysis can be improved, and the collection can be effectively carried out.

The invention provides an alarm analysis method and device, and the method comprises the steps: determining a call chain corresponding to a transaction alarm under the condition that the transaction alarm of a system is monitored; analyzing the transaction alarm based on the abnormal index information of the call chain, and determining whether the alarm type of the transaction alarm is an abnormal root cause alarm of the call chain; if so, sending alarm information corresponding to the transaction alarm to operation and maintenance personnel; and if not, determining whether the alarm frequency that the system continuously generates transaction alarms is greater than or equal to an alarm threshold value, and if so, sending alarm information corresponding to the transaction alarms to operation and maintenance personnel. By analyzing the transaction alarms, when the alarm type of the transaction alarms is the preset alarm type or the alarm frequency of the transaction alarms continuously generated by the system is greater than or equal to the alarm threshold value, the alarm information is sent to the operation and maintenance personnel, so that the alarm frequency of the system is reduced, and the attention and workload of the operation and maintenance personnel to the system are reduced; and the working efficiency of operation and maintenance personnel is improved.

The invention discloses an ICD (interface control document) designing method independent of a network. The ICD designing method independent of the network includes the following steps: a step 1: acquiring packaging rules of various network buses needing to be used; a step 2: inputting parameters and attribute information which need to be transmitted by a device to an application layer ICD; a step3: selecting a network bus needing to be used, and using the packaging rule corresponding to the network bus to define the application layer ICD; and a step 4: generating a physical layer ICD file. The ICD designing method independent of the network starts with the device application layer, analyzes differences of various network bus protocols, and forms the packaging rules of various network buses, can automatically package the application layer ICD through a software tool, and can perform automatic conversion to acquire the physical layer ICD. The method can improve the usability of device interface data between cross-project types, can shorten the ICD designing time, and can greatly enhance the readability of a device interaction relation.

The invention discloses a late sleep termination control method and system, which is suitable for a mobile device. The method comprises the following steps of identifying an application type of an executive application through a processing module, and according to the application type, providing breakpoint information of the executive application; and when the processing module judges that an executive state of the executive application meets at least one specific condition in the breakpoint information, providing caution information through a caution module. The method and the system are mainly used for reminding a user to sleep in a manner of not disturbing normal use of the application by the user in a sleep time period of the user, so that the antipathy degree of the user is reduced and the user experience is improved.

The invention discloses a course field-oriented image text aggregation method and system, and the method comprises the steps of taking object features and an adjacent matrix as input to construct an object relation graph corresponding to an image, and updating the features of nodes in the relation graph through a graph convolutional neural network; taking a set of feature vectors of all the language chunks as feature representation of the whole text description; taking the object feature set and the step language block feature set as input, and constructing a local similarity matrix between the image-text pairs; calculating the global similarity between the image and the text description in the whole course field; and training parameters of the image-text matching model through a gradient descent method, obtaining a section of text description with the highest global similarity through the learned parameters, and taking the text description as a description text matched with the image, so as to realize image text aggregation. According to the invention, the features of image and text description in the course field can be effectively extracted, so that schematic diagrams and text knowledge fragments in the course field are aggregated, and cross-modal knowledge fragments are automatically constructed.

The invention provides a method for realizing a quasi-real-time status service under a low-flow condition, which comprises the following steps of: judging the status of a mobile terminal according to the acquired characteristic value of a mobile terminal, wherein the status of the mobile terminal is online or offline; determining parameters for pushing information to the mobile terminal according to the status of the mobile terminal; and pushing the information to the mobile terminal according to the parameters. Correspondingly, the invention also provides a server for the method. According to the method provided by the invention, the real-time requirements of various applications can be ensured under the low-network flow condition.

The embodiment of the invention discloses an information processing method and electronic equipment. The method comprises the steps of determining a display position of to-be-displayed content; underthe condition that the display position of the to-be-displayed content at least indicates an edge region formed along the inner contour of the display output region of the display screen, when the to-be-displayed area is displayed, the edge area presenting different display effects relative to a non-edge area in the display output area; Therefore, the attention of a user to an area defined by theinner contour of the display output area can be reduced, the abrupt feeling of the area defined by the inner contour of the display output area relative to the display output area is reduced, and theuser experience is improved.

The invention relates to the technical field of advertising boards, in particular to a multifunctional inkjet advertising board, which comprises a base, a supporting column, a receding groove, a rack,a fixing plate, a driving element, a connecting frame, a rotating shaft and a gear. The advertising board has the beneficial effects that the driving element is arranged to drive the connecting frameto vertically move; the gear is in engaged transmission with the rack; furthermore, the rotating shaft is driven to rotate, a flex banner mounting panel is driven to turn over; in this way, in rainydays, the flex banner mounting panel can be turned over by 90 degrees; the flex banner mounting panel can play a role in sheltering from rain; in this way, pedestrians can stand below the flex bannerinstallation panel to shelter from rain, in the rain sheltering process, the contact time with the advertising board is prolonged, the advertising effect of the advertising board is facilitated, in addition, due to the fact that the advertising board can be turned over, the attention of the pedestrians is additionally increased, the attention is improved, and the advertising effect is improved.

The present invention relates to operating a biopsy unit. In order to provide an enhanced and facilitated way of controlling a biopsy unit, a control device (10) for controlling a biopsy unit is provided that comprises a support structure (12) with a housing (14), and a user interface unit (16) with a plurality of control elements (18). The control elements are configured to control the movement of a biopsy needle device along at least three moving direction lines. At least two of the moving direction lines are aligned to axes of a Cartesian coordinate system (20) and one moving direction line is aligned to a needle axis direction (22) of an elongated needle device of the biopsy unit, the needle axis direction being inclined to at least one of the axes of the Cartesian coordinate system. For each moving direction line, the housing is provided with a surface portion (24, 26, 28) that is aligned with a respective one of the moving direction lines. The control element for each moving direction line is arranged on the housing on a respective one of the surface portions; the control element for the movement along the needle axis direction being provided on an inclined surface portion (28) that is aligned with the inclined needle axis direction.