Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

115results about How to "Correct mistakes" patented technology

Method and system for generation and playback of supplemented videos

The present invention is a method and system for generation and playback of supplemented videos which include interactive features. The supplemented video includes hotspots that allow a video viewer to interact with the video when the video viewer sees an object. The hotspots can be manually defined. Information regarding the object and the hotspot can be stored in a separate XML file. Furthermore, the present invention can be a marketplace where a desired object can be found by searching the XML file. The search results can list the supplemented videos which contain hotspots corresponding to the object and also the time in the supplemented video in which the object is found. The present invention can also aggregate data about the objects based on the playback of the supplemented videos and the video viewer's interaction with the supplemented videos.
Owner:AIBUY INC

Medical record electronic data identification method and system thereof

The present invention discloses a medical record electronic data identification method and a system thereof. When a medical record electronic data identification mode is applied, through establishing a medical record template, the check results corresponding to key medical vocabulary information in a medical record image is put into the medical record template, since the medical record template comprises the key medical vocabulary information, the check result is put into the frame of the key medical vocabulary information in the medical record template, medical record electronic data is outputted, and the speed and accuracy of identification are improved at the same time.
Owner:SHANGHAI YIYI INFORMATION TECH CO LTD

Navigation system user voice custom method

A navigation system user voice custom method comprises hardware of a location module, a calculation way module, a guide module, a target tracking system (TTS) voice synthesis module and a decoding module. When the guide module generates character strings in buffer memory, voice in the buffer memory can be played directly. Before voice is synthesized, a system can compare character strings of guide information with all character strings in the buffer memory to play voice. A custom voice file is of a structure identical with a structure of a voice cache file, the custom voice file is formed by an index and a data portion, and the voice cache file is also formed by an index and a data portion. Personal computer (PC) terminal software can firstly read the voice cache file, the custom voice file is added to navigation equipment, and a navigation system starts the custom voice function. A voice recording tool can collect and store voice into a data file, and the voice is stored is guided into the navigation system to serve as a custom voice library. The navigation system user voice custom method has the advantages of using the voice custom function, correcting error of synthetizing voice with a TTS engine, adding custom capacity of the navigation system and quickening real time performance of voice prompt.
Owner:SHENYANG MXNAVI CO LTD

Hand gesture recognition method based on multi-feature fusion and fingertip detecting

ActiveCN104299004AThe amount of feature calculation is smallCorrect mistakesCharacter and pattern recognitionFingertip detectionSupport vector machine
The invention discloses a hand gesture recognition method based on multi-feature fusion and fingertip detecting. The method comprises a training process and a recognition process. In the training process, for a complex hand gesture, reasonable hand gesture features are selected, a multi-feature fusion feature extracting algorithm is used, the hand gesture is subjected to support vector machine training, and a training model is formed. In the recognition process, for an input video image sequence, hand gesture detecting is carried out first, then multi-feature extracting and fusion are carried out, and multiple features are input into the support vector machine to obtain a recognition results. Meanwhile, the hand gesture is subjected to fingertip detecting based on defects, through a defect screener, the positions of fingertips of fingers are located, then two-time recognition and detecting results are subjected to synthesized, and the final hand gesture recognition results are obtained. The problem that in a complex scene, the hand gesture recognition rate is not high can be effectively solved, the requirement of real-time performance is met, and the method can be well used in human-machine interaction.
Owner:ZHEJIANG UNIV

Double-frequency simulation coding light fault-tolerant phase expansion engineering method

InactiveCN104374337AAvoid Gross Measurement ErrorsCorrect mistakesUsing optical meansInformation extractionAlgorithm
The invention provides a double-frequency simulation coding light fault-tolerant phase expansion engineering method. According to the method, double-frequency simulation codes are composed of a simulation code 1 and a simulation code 2, the cycle length a1 of the simulation code 1 and the cycle length a2 of the simulation code 2 are set to be positive integers, a2 is equal to a1+1, and calculating is facilitated; according to the relation, a1 and a2 are zoomed with the same proportion, and the double-frequency simulation codes with any cycle length can be achieved. In the phase expanding process, a fault-tolerant engineering model is utilized for obtaining a phase expansion value, and the influence on the phase expansion value from the most error-prone simulation code cycle ordinal number is avoided. In the whole measuring space, errors, caused by errors of the simulation code cycle ordinal number, of the phase expansion value are corrected, measurement gross errors are avoided, and namely, the measurement gross errors caused by image information extraction errors are avoided.
Owner:HARBIN UNIV OF SCI & TECH

Multi-hypothesis map matching method based on vehicle state transition

The invention discloses a multi-hypothesis map matching method based on vehicle state transition, comprising the following steps of: S1, obtaining GPS (global positioning system) data and INS (inertial navigation system) data; S2, defining a multi-hypothesis tree according to the GPS data and the INS data, wherein the multi-hypothesis tree is used for recording all traveled paths of a vehicle, a root node is in the multi-hypothesis tree, and the root node represents a start point of the vehicle; S3, generating new hypothesis nodes in the multi-hypothesis tree when the vehicle obtains the GPS data each time; S4, calculating weight of the hypothesis nodes according to the GPS data and the INS data, wherein the weight indicates the similarity of the information included in the hypothesis nodes to the real information; S5, deleting some hypothesis nodes according to the weight of the hypothesis nodes; S6, selecting the hypothesis node with largest weight as the optimal hypothesis node. The multi-hypothesis map matching method based on the vehicle state transition can record the continuous traveled path of the vehicle, and also can get better matching result under the condition that GPS signal quality is not good or the vehicle is in a region where roads are intensive.
Owner:WUXI RES INST OF APPLIED TECH TSINGHUA UNIV +1

Electronic organ remote intelligent teaching system

The invention provides an electronic organ remote intelligent teaching system and relates to the field of internet-of-things intelligent teaching. The electronic organ remote intelligent teaching system aims to solve the problems that the conventional remote electronic organ teaching system cannot provide big data reference for learners and the learners cannot independently select learning modes. A human-computer interaction subsystem inputs an instruction, acquires video and audio signals, converts the received signals and indicates whether keys of the electronic organ are pressed down or not; a key detection and control subsystem acquires key information and controls the electronic organ to generate sound; an automatic play subsystem controls the electronic organ to play; a network communication module transmits the key information and the video and audio signals; a user side main controller receives a user instruction, transmits the key information, transmits the signals to the human-computer interaction subsystem, and transmits the video and audio signals and the key information to the network communication module; a cloud server module exchanges the key information and the video and audio signals to store through a database module. The electronic organ remote intelligent teaching system has the beneficial effects that the learners can independently sect the learning modes and big data is provided.
Owner:HARBIN INST OF TECH

Navigation method, navigation apparatus, wearable device and storage medium

The embodiment of the invention discloses a navigation method, a navigation apparatus, a wearable device and a storage medium. The method comprises the following steps: when a navigation control instruction is detected, acquiring real-time image data acquired by a camera, wherein the camera is arranged on the wearable device, and the wearable device comprises a pair of smart glasses; determining position information of the current user according to the real-time image data and map information, wherein the position information comprises orientation of the user; and judging whether the positioninformation satisfies a correct navigation execution condition, if not, triggering a path deviation reminding event. With the method, the apparatus, the device and the storage medium provided in the scheme of the invention, navigation accuracy is improved, and errors of the route can be corrected timely.
Owner:GUANGDONG OPPO MOBILE TELECOMM CORP LTD

Edible mushroom environment big data detection system

ActiveCN112906735AFast learningImprove the ability to process uncertain informationMeasurement devicesCharacter and pattern recognitionEdible mushroomEngineering
The invention discloses an edible fungus environment big data detection system, which comprises an edible fungus environment parameter collection and control platform and an edible fungus cultivation environment big data processing subsystem, and is characterized in that the edible fungus environment parameter collection and control platform realizes detection, adjustment and monitoring of edible fungus environment parameters; the edible mushroom environment parameter big data processing subsystem comprises a parameter detection unit and an evaluation unit and is used for evaluating the edible mushroom environment; the method effectively solves the problems that an existing edible mushroom environment has no influence on the yield of the edible mushroom environment according to nonlinearity, large lag, large and complex area of the edible mushroom environment and the like of the change of the edible mushroom environment parameters, and does not predict the yield of the edible mushroom environment and accurately detect and adjust the edible mushroom environment parameters; therefore, the environmental yield prediction and the production management of the edible mushrooms are greatly influenced.
Owner:HUAIYIN INSTITUTE OF TECHNOLOGY

Method for counting sheets and separating waste in production of banknotes

A method for separating a defective piece of paper money from a complete page of paper money pieces in the procedure of printing paper money includes such steps as using a page checking machine to find out the defective piece of paper money from a complete page of paper money pieces on which the pattern has been printed, spraying fluorescent powder on it, detecting it by fluorescent detector, sending its serial number to number-printing computer, separating out the defective piece paper money, using a good one to replace it, and checking. Its advantages are high automation level and high correctness.
Owner:南昌印钞厂
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products