56 results about "Interaction field" patented technology

The invention provides a method and a system for eliminating multi-channel acoustic echo of remote voice frequency interaction. The method for eliminating the multi-channel acoustic echo of the remote voice frequency interaction comprises the following steps of: obtaining input multi-channel sound source acoustic signals; carrying out acoustic separation processing for the input multi-channel sound source acoustic signals; carrying out acoustic echo elimination processing for each-channel acoustic signals which are subjected to the acoustic separation processing through a self-adapting filter; combining and carrying out acoustic combination for the each-channel acoustic signals which are subjected to the acoustic echo elimination processing; outputting the acoustic signal which is subjected to the acoustic combination. The system comprises a DSP (Digital Signal Processor), a multi-channel acoustic separation combining device and a self-adapting filter module. According to the method and the system, provided by the invention, the multi-channel acoustic echo can be eliminated with lower cost, high-quality voice communication is supplied, the method and the system which are provided by the invention can be applied to various complicated remote voice bidirectional voice interaction fields, the smearing time for processing the echo can be increased to be 500 ms, the acoustic bandwidth is in the range of 20 Hz to 20 KHz, and the echo elimination ability is increased to be larger than 100 dB.

The invention relates to the technical field of natural language processing, the artificial intelligence field and the human-computer interaction field, and discloses an intention identification method and a device capable of on-line error correction updating, which solve the problem that the existing intention identification model cannot be on-line error correction updating and can ot meet the requirement of user differentiated intention identification. In this invention, At first, that input text is preliminarily identify, if the intent identification result does not meet the user 's requirements, the preset intention category tag feedback is presented to the user, combining user selection of intention category labels and corresponding input data into an on-line error correction update training sample, A user online error correction update training data set is generated according to the online error correction update training sample and the training data set, A user online error correction update intention recognition model is obtained by training a user online error correction update training data set with an intention recognition basic model, Thus, the online error correction updating of the intention recognition model according to the user can be realized, the accuracy of the online error correction updating of the intention recognition model identification of the user ishigher, the error correction updating requirement of the user can be responded more quickly, and the satisfaction degree of the user can be improved.

The invention discloses a real-time gesture recognition method based on finger division and relates to a novel gesture recognition technology. The gesture recognition technology is one of hot topics in the human-computer interaction field. The real-time gesture recognition method based on finger division comprises the steps of extracting the whole hand region through a background subtraction method, and dividing a palm part and a finger part on the extracted hand region; judging finger types according to positions, angles and other information of fingers; performing gesture recognition on the whole gesture through a rule classifier. Related gesture recognition tests performed through a large number of experimental images prove that the real-time gesture recognition method based on finger division is high in speed and efficiency, and real-time effect can be achieved.

The invention provides a flexible object deformation simulation method for force haptic human-computer interaction. Based on a spring mass point model for equilateral triangle segmentation, according to the deformation of a flexible object, acting force is sequentially spread to all mass points in the structure layer by layer from the surface mass points inside a local area until reaching a certain preset boundary; overall consideration does not need to be conducted on the deformed object, and calculation resources are effectively and intensively put into the mass points in the local area. According to the flexible object deformation simulation method, only the mass points in the local area on the surface of the flexible object are involved, the deformation area is determined by contact force, the segmentation precision can be adjusted based on accuracy requirements, calculation quantity is greatly simplified, the deformation calculation speed is improved, adjustment is flexible, the application range is wide, and the flexible object deformation simulation method can be applied in the deformation situation where isotropous flexible objects are in stable contact, such as virtual surgical operation simulation and teleoperation robot control virtual reality human-computer interaction fields.

The invention discloses a dragging method based on a visualized layout system. The method comprises the steps that a target control to be dragged is selected, and the type of the target control is recorded; the target control is dragged to a target position, and a pop-up box is set according to the type of the target control and records database fields; the pop-up box binds the database fields with the target control, and the target control interacts with a background database according to the database fields and counts interaction fields; the counted interaction fields are returned back to the target control and displayed on the target control. By means of the dragging method based on the visualized layout system, a webpage can be quickly established.

The invention discloses a page ordering method and device for an operation interface and relates to the interaction field. The page ordering method and device for the operation interface are used for providing a more flexible page ordering mode so as to reduce the misoperation of a user. The page ordering method for the operation interface includes that acquiring click operations for thumbnails when locating at the page ordering mode, recording the triggering time of the click operations in sequence, wherein the thumbnails are in one-to-one correspondence to the pages on the current operation interface; confirming the show order of all the pages on the current operation interface according to the triggering time and clicked thumbnails according to a preset rule; ordering all the pages according to the confirmed show order. By means of said mode, the page ordering method and device for the operation interface realize to provide the more flexible page ordering mode so as to reduce the misoperation of the user.

The invention relates to an intelligent information interaction method, a system and a device for assisting vehicle running, belonging to the information interaction field. At present, there are someproblems such as single mode of interaction and inaccurate information between automobile and people, between automobile and road facilities. A nework composed of a vehicle-computer equipment, a relaypoint and a receiving terminal as nodes is desigen. The vehicle-computer equipment installed on the vehicle is only used for collecting and transmitting information, each vehicle-computer sends the optimized information to the terminal through the transfer point, and the terminal processes the information and uploads the information to the network of the cloud server, so that the driver can obtain other vehicle information visible and invisible from the network through the vehicle machine and mobile phone in real time to perceive the driving state of himself or others, and take countermeasures. The method of the invention detects by various sensors, the transmission of the information is not affected by the line of sight obstruction and bad weather, the data can be transmitted in real time, and the accuracy of the data is ensured.

Workflow sharing system and method are disclosed that allow a workflow server to provide a shared workflow, wherein the shared workflow generates at least one manifest, wherein the at least one manifest comprises a third-party interaction field, a credentials field, an input parameter field, and an output parameter field, wherein the workflow server receives a request for access to the shared workflow and transmits the at least one manifest of the shared workflow in response to the request for access to the shared workflow, wherein the at least one manifest comprises an acceptance requirement of the third-party interaction field, the credentials field, the input parameter field, and the output parameter field, wherein the workflow server receive consents of the acceptance requirement of the at least one manifest; and allows access to the shared workflow in response to the received consent of the acceptance requirement of the at least one manifest.

A permanent magnet rotational motor employs the use of a rotating iron band member which absorbs the magnetic flux of both of the opposing magnet flux fields from stator and rotor magnets. This causes mutually opposing, interacting magnetic flux fields of the rotor and stator magnets to enter in close proximity to one another without opposition. This results in the two interacting fields remaining totally one within the other, until the iron band member between the two interacting fields is rotably extracted, compelling the two interacting opposing flux fields to repel each other and causing the rotor assembly to push the power assembly crankshaft in the desired direction. In this manner the aforementioned repelling and attracting effects between the interacting fields experienced in the prior art, which negates rotation in permanent magnet motors, is eliminated. The present invention also provides a device to maintain the necessary coactive relationship between the opposing magnetic fields of rotor and stator magnets throughout a complete rotational cycle of the permanent magnet motor.

The invention discloses a node contact generation method and device for a mind map, an apparatus and a storage medium, which relate to the interaction field. The invention specifically confirms a target node and receives relevant instructions generated by confirming the contact. The starting point and the ending point are confirmed according to the target node, and the contact line based on the starting point and the ending point is established. A contact description box is displayed on the contact line; the contact description box receives and displays a contact description for describing therelationship between the first target node and the second target node, realizes the cross-association between the nodes of the mind map, and enhances the establishment and expression of multiple relationships among the information in the mind map.

The invention relates to the human-computer interaction field, in particular to an information input method based on a broken line gesture. The information input method based on the broken line gesture includes: classifying broken lines according to direction angles of directed line segments of the broken lines; building a corresponding relationship between each broken line type and a specific character or instruction; detecting a broken line movement track input by a user through a sensor or a sensor group; judging types of the broken lines according to the direction angles of all the line segments of the broken lines; confirming the characters or the instructions, respectively corresponding to the broken lines, according to the types of the broken lines; inputting the corresponding characters or executing the corresponding instructions. The information input method based on the broken line gesture only needs the user to input the broken line gesture so as to input the characters, and is suitable for inputting the characters or the instructions on a small screen.

The invention provides a vehicle-mounted HUD man-machine interaction system based on gesture recognition and belongs to the man-machine interaction field. The system comprises a gesture information acquisition module used for acquiring a sequence frame carrying gesture information of a driver, a data processing and recognition module used for processing the gesture information into a gesture instruction, and a vehicle-mounted HUD module used for executing the gesture instruction, wherein the vehicle-mounted HUD module displays the operation information on an object in the driving visual fieldof the driver. The gesture recognition technology and the vehicle-mounted HUD are combined, the interaction mode of a traditional vehicle-mounted HUD is changed, the existing problem of distraction ofdriving is solved, and therefore a safe driving environment is created.

The invention relates to the technical field of e-commerce cloud computing, in particular to an application loading method based on e-commerce cloud computing and an artificial intelligence computingcenter. According to the method, application loading data is divided into a series of relational network circle data in advance, and then demand trajectory field analysis is carried out on each pieceof relational network circle data through a target artificial intelligence calculation model to identify a demand trajectory field type to which each piece of relational network circle data belongs; and thus, the big data acquisition strategy updating accuracy is improved. After that, a demand trajectory field analysis result is further obtained, so that when it is detected that the relational network circle data meeting the preset service demand condition exists in the relational network circle data, it is rapidly determined that the target application loading object may be a big data collection service interaction field carrying key application loading interaction, and after the targeted strategy is updated, the subsequent low-confidence big data acquisition data volume is reduced, and the big data acquisition pertinence is improved.

The present invention relates to a power generation method; in particular, to a semiconductor ion energy power generation method, specifically: ionic liquid is placed in a container, and then at least one group of power generation units are built into the ionic liquid, and each power generation unit is separated by a partition wall. The power generation unit comprises at least one pair of P-type semiconductor slices and N-type semiconductor slices arranged in pairs, a partition is arranged between the P-type semiconductor slices and the N-type semiconductor slices, and the electric field of the P-type semiconductor slices and the N-type semiconductor slices is in contact with the ionic liquid. The ion-electric field interaction of the ionic liquid, the positively charged ions in the ionic liquid flow to the N-type semiconductor sheet under the action of the electric field, and the negatively charged ions in the ionic liquid flow to the P-type semiconductor sheet under the action of the electric field. The type semiconductor chip is connected to the conductive terminal. The invention does not consume extra energy, does not pollute the environment, does not generate noise, and can generate electricity continuously.

The embodiment of the invention discloses a node adding method, a device, a device and a storage medium of a mind map, which relate to the interaction field. The invention specifically generates an instruction by receiving a node, and the node generates an instruction to designate a target node. Recognizing the boundary of the display area of mind map; A new child node is added for the target nodewithin the boundary of the display area, the position of the target node in the display area is unchanged, and the new child node is located in the peripheral associated area of the target node. In the editing process of mind map, all the graphic elements of mind map are displayed in the scope of screen, which is convenient to view the whole content of mind map, and the interaction is more convenient, at the same time, the utilization rate of screen space is also improved.

The invention relates to the intelligent interaction field, in particular to a terminal control system and an operation method thereof. A mobile terminal software control way is set at a cloud platform, so that stability of the mobile terminal as well as the built-in software of the mobile terminal can be controlled simultaneously and the two kinds of effects can be realized without more operations. Therefore, the system and method are convenient and practical.

The invention relates to a biological information technology, and provides a method for accurately and efficiently predicting the interaction between a protein and a protein, capable of improving theeffect of useful information in an amino acid sequence in the prediction operation, and at the same time effectively reducing the influence of useless noise information. The protein-protein interaction prediction method using multivariate mutual information and residue binding energy includes the steps: (1) grouping amino acid categories; (2) defining feature representation; (3) establishing a feature frequency table; (4) calculating mutual information features; (5) calculating 3-element group mutual information features; (6) calculating amino acid physicochemical property features; (7) calculating an amino acid contact matrix AAC; (8) extracting amino acid sequence features; (9): performing singular value decomposition; and (10) obtaining an interaction between two proteins. The protein-protein interaction prediction method using multivariate mutual information and residue binding energy is mainly applied to the field of prediction of interactions between proteins.

The invention discloses a virtual touch screen human-computer interaction method, system and device based on a depth sensor. The system comprises a construction module, a division module, a calculation module and a control module. The method comprises the steps that a virtual touch screen interactive space is constructed according to virtual touch screen anchor points; the virtual touch screen interactive space is divided; interactive touch points are monitored, and touch force factors corresponding to the interactive touch points are calculated; and operation control is performed according totwo-dimensional coordinate positions and the touch force factors corresponding to the interactive touch points. The device comprises a memory and a processor used for implementing the virtual touch screen human-computer interaction method. By use of the virtual touch screen human-computer interaction method, system and device, the reliability and flexibility of virtual touch screen interaction can be greatly improved, responses can be made to different touch screen operation based on region boundaries, misoperation is reduced, and the accuracy of virtual touch screen operation is greatly improved. The virtual touch screen human-computer interaction method, system and device based on the depth sensor can be widely applied to the human-computer interaction field.

Disclosed is a method for the detection of an interaction between substances. According to the method, a salt-containing HEPES buffer is allowed to exist in a reaction region that provides a place of interaction for the interaction. The concentration of the salt in the salt-containing HEPES buffer may be adjusted such that one or more substances each having a negative charge, such as a probe nucleic acid, target nucleic acid and/or intercalator, can be prevented from undergoing non-specific adsorption on a positively-charged surface of a solid phase. Also disclosed are an interaction detecting unit including the reaction region and a medium existing in the reaction region and including the salt-containing HEPES buffer; a bioassay plate including a DNA chip provided with a number of such interaction detecting units; a system for detecting an interaction such as hybridization; and a reagent kit including the salt-containing HEPES buffer.

The invention relates to the human-computer interaction field and discloses an information processing method and electronic devices so as to solve the technical problem of the low interaction efficiency between two electronic devices in the prior art. The information processing method is applied to the first electronic device and comprises determining a communication identifier of the second electronic device when or after the communication between the first electronic device and the second electronic device is established and adding a second user corresponding to the communication identifier of the second electronic device to a number station of the first electronic device so as to establish an interface for information interaction between a first user corresponding to the first electronic device and the second user. According to the information processing method and the electronic devices, the technical effect that the interaction efficiency between the two electronic devices is improved is achieved.

The invention provides a multi-modal gesture recognition and visual feedback mechanism combined method, which comprises the following steps of: by virtue of a visual sensor high-definition visible light camera worn on the head, a myoelectricity electrode worn on the upper arm of a user and a motion gesture data glove worn on the hand, carrying out multi-modal gesture recognition on the user; multi-source data such as visual images/videos, instantaneous high-density electromyographic signals and motion signals are collected, and then a multi-mode gesture fusion recognition method is used for conducting comprehensive recognition on gesture actions of a user. Besides, a three-dimensional interaction feedback control model and a gesture interaction visual feedback mechanism model in the interaction field are designed, visual feedback of gesture interaction is achieved, a design principle, a model specification form and principle guidance application of visual feedback mode application are provided, and it is ensured that the visual feedback can play the maximum advantage.

The invention relates to a video interaction data processing method and a device, electronic equipment and a storage medium. The method comprises the steps of obtaining interaction data generated by an interaction operation implemented for a video tag, wherein the video tag correspondd to a target video displayed by a client and at least one data item; determining a target data item in the at least one data item, matching a plurality of interaction fields contained in the target data item with the interaction data, and updating field values of interaction content fields in the plurality of interaction fields according to the interaction data; and according to the at least one preset statistical dimension, performing statistics on a field value of the interaction content field in the at least one data item corresponding to the target video to obtain interaction statistical data, wherein the interaction statistical data is used for being displayed in the client.

The invention discloses a method and system for interactively and originally displaying advertising information. The method comprises the steps that host advertisement information is distributed by aserver end, the host advertisement information comprises interactive fields, host advertisement content information fields of host advertisement, and an accompanied advertisement information data structure; the accompanied advertisement information data structure comprises advertisement content information and displaying manners; a client receives host advertisement information, the host advertisement is rendered according to the host advertisement information; when there exists interaction between a user and the host advertisement, whether or not the interaction matches with the expanded limited interaction behavior of interaction field makers is judged by the client, and if yes, the accompanied advertisement is rendered. By means of the method and the system, the problems that current advertisement products are limited by loss rate generated by interaction manners and are hard to convert can be solved, and advertisement are supported to be interacted with users in a more diversifiedmanner.