44results about How to "Real-time conversion" patented technology

The invention discloses a voice conversion system and a voice conversion method. The voice conversion system comprises a recording module, a voice recognition module, a voice synthesis module, a loud-speaking module and a user control module, wherein the recording module is connected with the voice recognition module so as to receive speaking voice of a speaker and record a voice frequency; the voice recognition module performs voice recognition according to the difference of the modes, extracts relevant voice feature information in a training mode and converts the voice signal into relevant character information in a conversion mode; the voice synthesis module receives and stores the voice feature information transmitted by the voice recognition model in the training mode, and synthesizing the received text information and the voice feature information of a special person in the conversion mode, and transmitting the synthesized voice to the loud-speaking module; a user can control a module control system to be started and closed, select a working mode and select an appointer to be synthesized, i.e. convert the sound into the voice of anyone.

The invention provides a graphic conversion method, a mobile terminal and a OCR server, wherein, the method includes the following steps: the digital camera integrated in the mobile terminal shoots the graphic information required; the mobile terminal transmits the graphic information to the OCR server; the OCR server analyzes the graphic information and completes the graphic conversion; and the mobile terminal receives a character string and the data relevant to the character string converted. The invention is applicable to a variety of special (for example, languages of any countries or irregular scheduling, etc.), and complex graphics or graphics including a large amount of data to realize precise and rapid conversion or translation. The flexibility and accuracy of the method facilitates users to capture the favorite pictures to carry out the real-time conversion or translation, thus greatly enriching the lives of users.

The invention discloses a style-controllable image text real-time translation and conversion method. The method comprises the following steps: taking a scene image as input; performing feature extraction by using a multi-layer CNN network, and detecting the position and form information of the image text; and then erasing text pixels based on a text positioning box to obtain a background image anda mask, and carrying out background image restoration by using a thick restoration network and a thin restoration network based on a codec structure; performing form correction and style removal on the image text to obtain a common font image text; and recognizing the image text by using a CRNN model, performing correcting by combining text semantics, and performing translating or converting according to requirements; performing stylization processing on the translated text by learning the artistic style of the original text; and outputting a scene image with a controllable text conversion style. According to the method, more valuable information can be analyzed from the scene image, and the information storage degree during image text translation and conversion is remarkably enhanced.

The invention relates to a technical scheme for transforming 2D video image signals to 3D video image signals and application thereof. A video image signal flow decomposing and combining module, a video image signal flow delaying module and a video image signal flow interface transforming module are combined in the technical scheme and keep electrical connection mutually. The inputted 2D video image signal flows in different interface formats are decomposed into two paths of video image signal flow, one path of video image signal flow is properly delayed, then the processed two paths of video image signal flow are sent to a display terminal for displaying, the video display terminal displays two slightly different images with parallax for the same image content at the moment, and then strong stereoscopic impression is presented when people watch through corresponding special three-dimensional glasses. The invention is widely applied to equipment comprising broadcast television networks, electronic video equipment, internet, DVD, VCD, computer videos and the like.

The invention discloses a real-time traffic condition information generating method which comprises the steps of: automatically receiving broadcast; performing speed recognition on a broadcast signal, performing keyword detection, extracting traffic condition information from the broadcast signal, and converting the traffic condition information to text information; converting the text information to a map feature language through natural language processing technique, drawing and feeding back in a map product for forming traffic condition information. Correspondingly, the invention further provides a real-time traffic condition generating system. The real-time traffic condition information generating method can provide real-time traffic condition information for users in a wide area. Based on the real-time traffic condition information, a series of applications of navigation, going-out planning, etc. can be performed. Furthermore a large amount of labor can be saved for a traffic condition product provider.

The invention discloses a hand five-finger rehabilitation training and correction device and relates to the technical field of rehabilitation devices. An upper box body and a lower box body are connected in a matched mode, a containing cavity is formed in the inside, a hand training mechanism is fixed to the upper box body, prime mover mechanisms are arranged in the containing cavity and transfer power to the hand training mechanism through a transmission mechanism, the hands are driven to perform rehabilitation training. The hand five-finger rehabilitation training and correction device is provided with two different prime mover mechanisms, adopts hand-cranking and motor hybrid drive and provides active, passive and cooperated training modes, the designed hand-cranking drive is used for healthy limb training and promoting rehabilitation of a hand nerve function and a double-side cooperated function, a more simplified parallel six-rod mechanism is adopted, and the actual holding rule of the hands is greatly recovered. Quick and convenient dismounting of a drive rod and adjustment of a drive rod installing position are achieved through exercise and limiting of a secondary slide block, thus the exercise angles of knuckles are adjusted in a certain range, and the hand five-finger rehabilitation training and correction device achieves the purposes of meeting different rehabilitation training demands and achieving personalized training.

The invention belongs to the field of the non-destructive testing technology for a nuclear power station and in particular relates to an eddy-signal-recognition-technology-based full-automatic data acquisition method. The eddy-signal-recognition-technology-based full-automatic data acquisition method comprises the following steps: 1. setting the detection boundary and the data recording boundary and building characteristic signal models of eddy digital signals at the boundary positions; 2. controlling a probe to enter a to-be-detected pipeline through a pipe opening and move inwards at designated speed; 3. acquiring an eddy digital signal in real time by an eddy current instrument, and when the acquired signal is matched with the characteristic signal for detecting the boundary, stopping advancing; 4. starting to acquire data automatically, enabling the probe to move back from the to-be-detected pipeline outwards after the probe reaches the detection boundary, simultaneously, starting to acquire the signal automatically until the probe reaches the recording boundary; and 5. stopping automatic data acquisition, after the probe reaches the recording boundary, stopping data acquisition, and enabling the probe to move back out of the to-be-detected pipeline. Compared with the method that a sensor is additionally arranged at the pipe end of a push-pull device, the signal-recognition-based positioning method can save the cost and has high reliability.

The invention relates to a variable lane real-time control system and method. The system comprises an equipment monitoring subsystem, a lane state studying and judging subsystem, a scheme studying andjudging subsystem and a scheme decision-making subsystem. The variable lane real-time control system and method not only solve a timing switching control mode of a traditional variable lane signal lamp/lane guide plate in the market, but also realize the self-adaptive change according to the real-time traffic flow condition of the intersection, the real-time road right distribution is more reasonable, the utilization rate of road resources is obviously improved, the traffic capacity of the intersection is obviously improved, the urban traffic jam condition can be better improved compared withan old-fashioned timing switching mode, and a wider application range is realized.

The invention discloses an intelligent loading and unloading vehicle system based on visual servo, and belongs to the technical field of intelligent loading and unloading vehicle robots. The intelligent loading and unloading vehicle system comprises an unstacking and stacking mechanism, a conveying mechanism and an AGV loading and unloading mechanism, wherein the two ends of the conveying mechanism are connected to the unstacking and stacking mechanism and the AGV loading and unloading mechanism correspondingly; the AGV loading and unloading mechanism comprises an AGV trolley capable of achieving front-back, left-right and rotation displacement, a pose adjusting device arranged on the AGV trolley, a flexible grabbing device arranged on the pose adjusting device, and a first image obtainingmechanism arranged on the flexible grabbing device; and the unstacking and stacking mechanism comprises a movable carrying platform, a three-coordinate machine arranged on the carrying platform, anda second image obtaining mechanism arranged on the three-coordinate machine. The intelligent loading and unloading vehicle system solves the problem that an unstacking and stacking machine is not suitable for narrow space and space needing to be moved, full-automatic loading and unloading of goods in limited space such as a van are achieved, and manpower replacement is achieved.

The invention discloses a television color temperature control method and system. The television color temperature control system comprises a memory, a processor and a user interface module. The user interface module is used for regulating white balance parameter values of multiple groups of color temperatures and converting the white balance parameter values into configuration files. The configuration files are stored in the memory through the processor. When the user interface module receives color temperature mode switching signals and correspondingly sends control signals, the processor is used for reading the configuration files from the memory and converting the configuration files into the white balance parameter values of the multiple groups of color temperatures so as to enable the user interface module to obtain the corresponding white balance parameter values from the white balance parameter values, and the white balance parameter values are used for adjusting current color temperature mode color temperatures. According to the television color temperature control method and system, the white balance parameter values are stored through the configuration files, operation is very flexible and concise, the white balance parameter values of the color temperatures can be adjusted through the user interface module, and convenience can be brought to system development and maintenance.

The invention discloses a real-time analyzing method of an SV (Sampling Value) message of an intelligent substation. The real-time analyzing method comprises the following steps of: A, grabbing the SV message sent by an IED (Intelligent Electronic Device); B, respectively obtaining Header MAC, Priority Tagged Ethertype, Header and MAC (Multi-Access Computer) computation information from the grabbed SV message; and C, analyzing the obtained data information. The real-time analyzing method of the SV message of the intelligent substation has the beneficial effects that the SV message sent by all IEDs can be analyzed; and the SV message parameters are analyzed to realize real-time conversion of the SV message. The converted SV message format is consistent with a remote measuring expression format of the conventional substation; and the specific information of SV is correctly indicated, so that the debugging efficiency is improved. Especially, when the message is analyzed, the site construction is not affected, and the normal communication of the devices is ensured.

The invention, which belongs to the 3D animation creation platform field, especially relates to a 3D animation film forced working creation platform. According to the scheme, display card outputting is carried out; synchronous conversion is carried out on a signal; and a three-dimensional pre-monitoring is completed based on a mapping relation of four screens. A signal conversion device is arranged inside the system; and signal conversion can be carried out by outputting of three display heads of the conventional display card so as to complete connection. On the basis of performance improvement of the traditional computer, browsing and generating of video and three-dimension files can be realized rapidly in real time and real-time conversion and operation can be realized. Therefore, the provided creation platform is a cross-platform and multi-compatible system capable of realizing real-time conversion and rapid operation.

The invention discloses an intelligent dual-step double-precision weighing sensor comprising a mounting seat, a first precision weighing sensor, a second precision weighing sensor, a supporting leg, and an elastic supporting element. The mounting seat is provided with a mounting plate part and a guiding and supporting part for transmitting pressures to the mounting plate part; and the elastic supporting element, the first precision weighing sensor, the mounting plate part, the second precision weighing sensor, and the supporting leg are arranged successively in a pressing mode. The elastic supporting element consists of an upper support element, a spring, and a lower support element; a guiding and pressing part is arranged on the bottom wall of the upper support element; and the guiding pressing part is higher than the top of the guiding supporting part in a free state. A static contact is arranged at the top of the guiding and supporting part; and a moving contact matching the static contact is arranged at the guiding and pressing part of the upper support element. When the guiding and pressing part presses the guiding and supporting part, the moving contact presses the static contact. The intelligent dual-step double-precision weighing sensor has the continuous dual-step ranges and has different precision values in ranges at all steps.

The invention provides a deaf-mute assistant guiding tool. The tool comprises a body, a display device, a functional button, a recording device, a voice signal processing module used for converting atext, a communication input device, a power module and a controller. The tool is characterized in that the power module is arranged inside the body and is connected with the display device, the recording device, the voice signal processing module, the communication input device and the controller respectively. According to the deaf-mute assistant guiding tool, the voice signal processing module isadopted to perform real-time conversion on a recorded speech, so that a produced sound better meets an actual expression requirement; and the communication input device is adopted to enable a personwith a voice disorder to perform speech input, and the speech is processed through the voice signal processing module and output in real time through a sound production device.

The invention relates to a device and method for changing plane images into three-dimensional videos in real time. The device mainly comprises a plane image input unit, a processing unit, a making unit and a broadcasting unit, wherein the plane image input unit completes the obtaining of the plane images, the processing unit analyzes the plane images into first image digital information, the making unit horizontally displaces the first image digital information and/or forming second image digital information after the lagged displaying of the first image digital information, and the broadcasting unit forms the three-dimensional videos through the simultaneous displaying of the first image digital information and the second image digital information. The invention also provides the method for changing the plane images into the three-dimensional videos in real time, which is matched with the device for changing the plane images into the three-dimensional videos in real time. Through the device and the method, free conversion of the plane images can be realized, and the plane images can be conveniently converted into the three-dimensional videos in real time. The switching between all function is simple and convenient, and the three-dimensional image effect is smooth.

The invention provides a physical game device and relates to the field of simulation devices. The physical game device comprises a scene stage, scene props and a lifting mechanism. Installation holes are formed in a bottom board of the scene stage, each scene prop comprises a body and an end, the end penetrates the corresponding installation hole to extend to the position under the scene stage, and the lifting mechanism is arranged under the scene stage and can drive the ends to ascend or descend so as to enable the bodies to be presented or hidden on the scene stage. The physical game device can realize free transform of game scenes, the requirement of game players for frequent scene selection is met, interestingness of games is improved, and the physical game device has the advantage that the game scenes do no need to be transformed manually in a time-consuming and laborsome manner as compared with existing physical game devices.

The invention relates to the technical field of accelerators, in particular to a system for accelerator excitation curve conversion. The system is used for converting a physically set magnetic field value into a current value required by a power supply or converting the current value of the power supply into a corresponding magnetic field value by an accelerator, and the system can be applied to a large science device, is low in equipment requirement, is high in compatibility, is simple to install and use, and can achieve the real-time conversion of a DC power supply, a pulse power supply with a preset value, a convex rail pulse power supply without a preset value, and a pulse power supply with a floating power supply at the same time.

The invention provides a real-time query system for organization data. The real-time query system comprises a database of at least one organization, a kettle, a data distributor, a Flink, an OpenApi and an external system, the database of each organization creates a view according to the business requirement of the organization, and data in the view is provided for an external system to inquire; when a query request sent by an external system is received, the kettle extracts data from a database view corresponding to the query request, and performs field conversion on the extracted data according to a preset field; the data distributor distributes the data subjected to field conversion to the Flink; the Flink performs format conversion on the distributed data according to a preset format,and sends the data subjected to format conversion to the OpenApi as a query result; the OpenApi outputs a query result to an external system; real-time conversion of organizational structure data is achieved, and an external system can inquire the organizational structure data in real time.

The invention discloses an adaptive protocol conversion system for high-voltage direct-current measurement and a using method thereof, and belongs to the technical field of power communication and automation. The adaptive protocol conversion system comprises a photoelectric conversion module, a protocol conversion unit and a communication interface module. The adaptive protocol conversion system can receive optical digital signals of various IEC60044-8 communication protocols such as standard FT3 and state grid FT3 data frame formats specified by IEC60044-8, and convert the optical digital signals into IEC61850-9-1, IEC61850-9-2LE and state grid IEC61850-9-2 network digital message signals meeting the IEC61850-9 standard in a high-speed, stable and real-time manner for optical digital protection, control, metering, measurement and test equipment. The protocol conversion system has the functions of intelligently identifying the Baud rate, the sampling rate, the link layer length and thenumber of check bytes of an input signal, has high-speed, stable and real-time conversion performance, and is widely applied to testing and verification of intelligent substations, converter stationmeasurement systems and digital equipment.