46results about How to "Reduce study load" patented technology

The invention provides an intelligent health management system based on an elder-caring robot; the system comprises an elder-caring robot, a remote terminal and a cloud server; the elder-caring robot comprises a reminder and a detector mutually connected; the reminder starts path planning and mobile functions according to a preset medical solution, searches the elder according to a face recognition technology and a sound source positioning technology, communicates with the elder through a voice interaction technology, and broadcasts prompt contents; the detector monitors an intelligent medicine box opening state on the elder-caring robot, detects the elder complete conditions on the prompt contents through the image identification technology and the voice interaction technology, and synchronizes the detection result onto the cloud server; the cloud server sends the information to the remote terminal, thus returning feedbacks to children or medical personnel. The intelligent health management system can provide complete, professional, intelligent and safe one-stop intelligent health management services, so children can remotely monitor, track and detect elder living conditions in a long term, thus realizing disease types and risk predictions.

The invention discloses a data transmission method and a data transmission system. The data transmission method comprises the following steps: when an MAC (media access control) agency module receives a message forwarded by a virtual switch, replacing a source MAC address in the message with a corresponding MAC address by the MAC agency module; forwarding the message in which the source MAC address is replaced to switching network equipment by the MAC agency module, and processing the message by the switching network equipment; when the MAC agency module receives the message forwarded by the switching network equipment, replacing a target MAC address in the message with a corresponding interior MAC address by the MAC agency module; and forwarding the message in which the target MAC address is replaced to the virtual switch by the MAC agency module, and forwarding the message to a target virtual server by the virtual switch. By the data transmission method, the consumption of the MAC address is reduced, system resources are saved, and the MAC address learning load of the switching network equipment is reduced.

The invention discloses an internet precision education tutoring system based on big data and artificial intelligence. The system comprises a teacher client, a student client, a parent client, an education manager client and a server. The teacher client, the student client, the parent client and the education manager client are connected with the server. School teaching and management are assistedbased on big data analysis and an artificial intelligence technology so that the education burden of teachers, society and education managers in three aspects is reduced. By establishing a digital resource library and careful knowledge graphs, learning conditions of students are accurately collected, and personalized tutoring plans of the students are formulated in an assisted manner. By establishing an interest stimulation module and a personalized evaluation system, the learning achievement and exploration spirit of students are stimulated. According to the invention, a four-in-one system of'school + teacher + student + parent 'is established in combination with the Internet, and the functions of real-time communication, mutual communication and self-promotion are realized.

The invention relates to an improved face key point detection method based on GIOU and weighted NMS. Firstly, a lightweight cascaded face target detection network is adopted to carry out detection androtation angle detection on a face target in an image, GIOU is adopted to replace IOU to serve as a target frame position precision index, and IOU loss is adopted to carry out position regression; secondly, a weighted NMS suppression method is adopted for the obtained face target, and a face target box with high confidence in the image is obtained; and finally, the face key points are detected and regressed by adopting a lightweight cascaded face key point detection network. Compared with a traditional convolutional network, the speed is improved, the real-time performance of the algorithm isenhanced, and the network detection precision is improved on the premise of not additionally increasing the operand. The problem that a traditional target detection and face key point detection method is poor in real-time performance is solved, and real-time key point detection can be carried out on face targets in multiple different rotating directions in a complex environment under the condition that the detection precision is guaranteed.

The invention provides a personalized practice method and system. The method herein comprises the steps of acquiring a user subject knowledge map, to be specific, acquiring the user subject knowledgemap from a knowledge map database according to a student's ID (identity) and subject; acquiring recent class information, to be specific, acquiring the recent class information from a class information database according to the student's ID and subject; generating a knowledge point practice list, to be specific, determining knowledge points to be practiced and their weights according to the recentclass information and the student subject knowledge map, and adding them to the knowledge point practice list; generating an exercise list, to be specific, generating the exercise list according to time to be spent for practice and the knowledge point practice list; evaluating the practice, to be specific, evaluating according to the student's exercising data, judging whether each exercise is correct, and updating related data; displaying practice results, to be specific, giving a summary to the practice. Practicing is performed personally according to students' degree of master for knowledgepoints through the above means, excessive exercises tactic is abandoned, and the student's learning efficiency and effect are improved.

The invention discloses a writing teaching interaction system by utilizing an internet and voice technology and a realization method thereof. The method comprises the following steps: uploading audio and video information or text content information when starting writing, wherein the audio and video information and / or text content information at least comprise(s) writing keywords; carrying out semantic analysis and matching archiving on the keywords, and establishing a composition practice electronic archive; correcting or commenting the composition practice electronic archive, and then, issuing feedback results; and repeating the steps above until the feedback results are marked as completion of writing. The system can effectively improve the efficiency of writing of students, reduce learning burden, improve teaching efficiency and reduce teaching time, and has extremely obvious effect especially for lower-grade students.

The invention provides a motor imagery brain-computer interface control method based on noninvasive electrical stimulation. The control method comprises the following steps: applying anode noninvasive brain function electrical stimulation on the primary motor cortex of a user; collecting an electroencephalography original signal after the user executes a motor imagery task, and extracting an event-related desynchronization signal and / or event-related synchronization signal from the electroencephalography original signal; carrying out quantification processing on the event-related desynchronization signal and / or event-related synchronization signal to obtain the average power of the event-related desynchronization signal and / or event-related synchronization signal in a Lower[Mu] waveband, an Upper[Mu] waveband and a [Beta] waveband; carrying out feature extraction on the event-related desynchronization signal and / or event-related synchronization signal to obtain the feature values of the event-related desynchronization signal and / or event-related synchronization signal; and according to the average power, classifying the feature values to obtain a control instruction, wherein the control instruction is used for controlling a feedback training normal form.

The invention relates to a medical sensor as well as a use method and operation equipment thereof. The medical sensor comprises a logo, a communication part and a setting part, wherein the logo can be used for showing the identity information of the medical sensor and can be identified by specific electronic equipment; the communication part can be used for receiving initialization parameters matched with the identity information from the electronic equipment; the setting part is communicated with the communication part and is used for finishing the initialization setting of the medical sensor according to the initialization parameters. The medical sensor is provided with the logo used for showing the identity information of the medical sensor, so that the operation equipment can obtain the identity information through the logo, the use information, such as the initialization parameters, of the medical sensor are read on the basis of the identity information to automatically finish the initialization setting of the medical sensor, a user can conveniently use the medical sensor, learning burden and operation complexity can be lightened and lowered, and working efficiency is improved.

The invention discloses an extension neural network pattern recognition method based on priori knowledge. The method includes the following steps that (1) a training sample set and a knowledge base are prepared; (2) an initial weight value of an extension neural network is determined according to training samples and the priori knowledge; (3) the extension neural network can be trained by the utilization of the training samples, if a training process is converged or the total error rate reaches a preset value, training is stopped, and a weight value vector, after the training, of the extension neural network is kept, and otherwise the training is continued; (4) the trained extension neural network is used for performing pattern recognition until recognition of all objects to be recognized is completed. According to the extension neural network pattern recognition method, under the common driving of the priori knowledge and the training samples, learning of the extension neural network is guided, training and learning of the extension neural network are completed, the learning burden of the extension neural network is relieved, the performance of the extension neural network is effectively improved, training time is shortened, and recognition accuracy is improved.

The invention discloses a many-to-many speaker conversion method based on Transitive STARGAN. According to the invention, a STARGAN generator is combined with a transmission network, features extracted by a coding network are transmitted to a corresponding network layer of a decoding network in the generator; the learning capability of the decoding network on semantic features of different scalesis improved; the learning function of the model on the frequency spectrum deep features is realized; the frequency spectrum generation quality of the decoding network is improved, semantic features and personalized features of speakers are learned more fully, so that the personality similarity and voice quality of converted and synthesized voice are better improved, the problem that the personality similarity and the naturalness are poor after STARGAN model conversion is solved, and high-quality many-to-many speaker conversion under a non-parallel text condition is realized.

The invention discloses a plane segmentation method based on superpixels and a graph convolutional network. The method comprises the following steps: preprocessing an input color image to obtain a preprocessed image; then, according to the image resolution, segmenting the preprocessed image into a proper number of superpixels, and converting the superpixels into an undirected graph structure; constructing a graph convolutional network, and training the graph convolutional network by using a data set; and finally, predicting a graph formed by the superpixels by using the trained graph convolutional network, and carrying out plane classification on each superpixel so as to complete plane segmentation. According to the method, the image is segmented into the superpixels, so that the edge information in the original image can be well reserved, the learning burden of a subsequent image neural network is reduced, and the problem that the segmented plane edge is too large in gap with the actual situation is prevented; and a specific label is extracted through a specific algorithm by utilizing an existing data set and serves as a training set for subsequent supervised neural network learning, so that the problem that the data set is not segmented for a plane is solved, and huge cost of manual marking is avoided.

A study method, study material label language and learning machine includes steps of selection, association, cooperation, velocity variation and repetition; the selection includes preselector and converging-diverging; the association includes catenation, looking up and regulation; the cooperation includes conversational, white plate, side-listening and checking; the velocity variation includes force-reading, guide-reading and flash-reading. The study material label language includes embedded multimedia synchronous element, embedded multimedia structure element, learner and study material interactive element, synchronous, interactive and structure are separable with resource. The learning machine includes learning subsystem, learning material label language subsystem, multimedia subsystem and communication subsystem. Selecting the proper study content randomly, associating and constructing new knowledge, enlightening and role acting, mobilizing, assisting, tracking, evaluating the thinking activity and knowledge structured and systematic of users, multimedia synchronizing listen-speak-read-write, learner and study material are interactive, from thickness to thinness, from slow to fast, understanding, remembering and training calculatingly. The learning efficiency is increased, learning berden is decreased and the learning interst is aroused.

The invention discloses an immersive remote interaction system and method based on natural walking. The system comprises a robot, a server, a location tracking device and a head-mounted virtual reality device worn on the head of a user in use; and the location tracking device and the head-mounted virtual reality device are in the same physical space, and the robot and the head-mounted virtual reality device are in different physical spaces. The action of the robot is remotely controlled according to action data of the user; and meanwhile, the action of the robot is transmitted to the head-mounted virtual reality device through the server, and then the action of the robot is displayed to the user, so that the interaction between the robot and the user is realized.

The invention relates to a dynamic vision sensor-oriented brain-like gesture sequence identification method. The method comprises the following steps of address event expression data capture, space-time convolution, pulse pooling, space-time feature learning and gesture sequence identification by using a learned DoubleSTS model. The method has the characteristics of high noise robustness, high precision, high efficiency, rapid convergence, time sensitivity, brain imitation and the like, and each gesture in the gesture sequence can be successfully recognized.

The invention discloses a flight mass spectrum biochip for career potential assessment. The biochip includes a sample analyte and a mass spectrum chip matrix that undergo cocrystallization. The sample analyte comprises a?target?gene, a PCR primer and a single base extension primer used for innate potential assessment. The invention also discloses a detection method making use of the chip. The method consists of: extracting the sample DNA of a detection subject; designing a PCR primer and a single base extension primer; conducting target gene PCR amplification and single base extension on the sample DNA to form a sample analyte;?and?subjecting the sample analyte and the mass spectrum chip?matrix?to cocrystallization, and detecting the SNP locus of the target gene. The gene chip provided in the invention can?get an in-depth understanding of the potential ability of a person and assess personal career potential?so as to combine personal development with organizational development to work out corresponding work, education and training action plans, thus making reasonable arrangement of the time, projects and measures for each step.

The invention discloses a video data feature extraction method based on audio and video multi-mode time sequence prediction, and the method comprises the steps: 1, obtaining a video data set through a video collection device, and constructing an audio and video double-flow data pair; 2, for each video frame in the video stream and each audio clip in the audio stream, adopting a series of data enhancement operations in respective modes, and converting a one-dimensional audio into a two-dimensional spectrogram; 3, constructing an audio and video multi-modal prediction model which comprises a video stream feature extraction network unit, an audio stream feature extraction network unit, a time sequence information aggregation network unit and a multi-modal interaction prediction network unit; and 4, according to the uncertain features obtained by the multi-modal interaction prediction, calculating the total loss of the multi-modal prediction of the audio and video and optimizing the network. According to the method, useful information of the video can be effectively mined in a self-supervised manner by utilizing the time sequence of the video and combining the interaction between audio and video double streams, so that the effectiveness of feature extraction is improved, and actual downstream tasks such as video understanding, sound source localization, anomaly detection and the like are facilitated.

A unit type curtain wall parametric modeling method based on BIM comprises the steps that a unit type curtain wall structure is decomposed and modeled according to a curtain wall design scheme or a CAD construction drawing and then parameterized combination is conducted, a unit type curtain wall BIM model is established by assembling parameterized curtain wall unit bodies, and information is transmitted through shared parameters; performing collision detection on the unit type curtain wall BIM model to realize model optimization, and generating a unit type curtain wall engineering scale and a curtain wall professional drawing by using a detailed statement function and a drawing drawing function in BIM software, so as to provide a basis for deep application of the unit type curtain wall BIM model; the method is low in operation difficulty, high in modeling efficiency, complete in model information and accurate in graph size, accurate information data is provided for unit type curtain wall design and construction and factory blanking production, and the professional efficiency of the curtain wall is improved.

The invention provides a small sample remote sensing ground object classification method and system based on a feature generation network. The method comprises the steps of dividing a remote sensing ground object classification data set into a source data set and a target data set based on ground object categories; training a feature extractor and a generation network on the source data set to complete parameter initialization of the feature extractor; on the basis of the target data set, using the feature extractor and the generation network for feature generation to obtain a generation feature set corresponding to each surface feature type, so that high-quality target data can be effectively expanded, and the over-fitting risk on the target data set is reduced; then, using a feature extractor for carrying out feature extraction on the test ground object image; and performing similarity measurement on the test sample features and the generated feature sets of the various ground objecttypes, giving confidence scores of the test images on the various ground object types, and further determining the ground object classification of the tested images, so the remote sensing ground object classification problem under the small sample condition is solved.

The invention discloses a gesture operation method and a multi-media playing device and is suitable for the multi-media playing device which has a touch control function. The gesture operation method comprises the steps of detecting gesture input while the multi-media playing device plays a video or a streaming video, and performing a corresponding function by the multi-media playing device when the gesture input comprises a specific number of finger touch control points which are carried out specific and relative movement along with time. Therefore, a user interface of the multi-media playing device enables a user to understand easily by employing the direct gesture input.

The invention relates to a Chinese character input method; the method can use a common English alphabetical keyboard or a touch screen virtual soft keyboard to input Chinese characters into electronic equipment like a computer, a flat computer and a mobile phone via various coding modes, thus carrying out Chinese and English information processing and communication; the method needs not to split Chinese characters, uses a locating code fetch mode to extract Chinese character font characteristic etymons and Hanzi pronunciation characteristics, can change Hanzi pronunciation coding element and font coding element combined components and code element positions, thus forming a Chinese character input integration system with various coding input modes; the touch screen virtual soft keyboard prompt function can realize binary syllabification vowel and coding etymon layout memory-free effect, thus solving the problems that a conventional pinyin input method cannot accurate type characters and a font input method is hard to memorize coding etymons, and forming a general Hanzi coding input method, wherein the novel method can use same coding rules to code a national standard GB2312-80 font library and UNICOD traditional font and simplified font large character set Chinese characters.

The invention relates to a communication tool for hearing-impaired people. The communication tool comprises a device main body and a wearable device which are in communication connection, wherein thedevice main body comprises a display module, an audio output module, a microphone and a processor; the processor is used for acquiring sign language action information of the hearing-impaired people through the wearable device, determining semantic information expressed by the hearing-impaired people according to the sign language action information, then converting the semantic information into first voice information, and controlling the audio output module to play the first voice information, and the processor is also used for acquiring second voice information of normal people through themicrophone, converting the second voice information into image information of sign language actions, and controlling the display module to display the image information. By implementing the technicalscheme, instant communication between the hearing-impaired people and the normal people can be realized.

The invention discloses an intelligent education system, which comprises a registration and login module, an interactive teaching module, a learning test module, an intelligent playback module and a data management module, and is characterized in that the registration and login module is used for a user to register an account and log in the education system, and the interactive teaching module is used for a teacher and students to carry out online interactive learning; the learning test module is used for evaluating the mastering ability of students on knowledge points, the intelligent playback module is used for intelligently playing back teaching videos, the data management module is used for managing user information and teaching resources, the user information comprises user types, user accounts, user names and user passwords, and the user types comprise students, parents and teachers; the teaching resources comprise live videos, playback videos, learning test exercises and student test scores, the system is scientific, reasonable and convenient to use, the knowledge point playback speed is configured according to the mastering condition of the students on the knowledge points, the students are helped to consolidate and review, and therefore the learning efficiency of the students is improved.

The invention discloses a language learning machine system based on a sharing mode. The language learning machine system comprises a language learning machine, a sharing site, a background server anda user mobile phone, wherein the language learning machine is in communication connection with the sharing site, the language learning machine is in communication connection with the background server, the user mobile phone is in communication connection with the sharing site, and the user mobile phone is in communication connection with the background server. The system can be used for shared learning, and the structure of the sharing system is simplified; the system is simple, the cost is reduced, and the learning efficiency of users is improved.

The invention discloses a teaching aid for simplifying a programming learning process on multiple aspects and a programming teaching aid. The teaching aid comprises a battery, a micro control unit, a program input assembly, an environment sensing assembly and a feedback assembly. The program input assembly is used for receiving program data or signals compiled by a user and used for controlling the feedback assembly to carry out corresponding actions; the environment sensing assembly is used for acquiring external environmental data, and transmitting the data to the micro control unit; the micro control unit is used for running a program input by the program input assembly, and generating a control instruction in combination with the environmental data received by the environment sensing assembly; and the feedback assembly is used for receiving and executing the control instruction. The aim of the invention is to simplify the programming learning process, simple contrast between the function and the program is realized by using the combination of the assemblies in the teaching aid, and the programming learning is simple and joyful.

The invention discloses an office area intelligent device management service system, and belongs to the intelligent device management field. The office area intelligent device management service system comprises an office area device management server, a wireless positioning and unvarnished transmission gateway, a mobile terminal and an office area intelligent device, the office area device management server is in communication connection with the office area intelligent device through a wifi wireless communication network, the office area device management server is in communication connection with the wireless positioning and unvarnished transmission gateway through the wifi wireless communication network, and the wireless positioning and unvarnished transmission gateway is in signal connection with the mobile terminal. The office area intelligent device management service system is high in intelligent degree, the office area intelligent devices are seamlessly connected through the device management server, the wireless positioning and unvarnished transmission gateway and the mobile terminal, the integrated monitoring, management and use of the office area intelligent devices aresimply and conveniently achieved, the operation process is simplified, and the user experience is improved.

The invention, which belongs to the technical field of education equipment, relates to an integrated knowledge retrieving machine for a school. The integrated knowledge retrieving machine comprises an equipment integrated bench; and a lower limb extending groove is formed in the front side of the equipment integrated bench. A fixed seat stand is arranged at the front side of the equipment integrated bench; and a flexible seat is arranged at the upper side of the fixed seat stand. A power line output port is formed in the left side of the equipment integrated bench; and a power lead is arranged at the left side of the power line output port and a power plug is arranged at the upper side of the power lead. A support rod is arranged at the left side of the equipment integrated bench; a waste paper basket is arranged at the upper side of the support rod; a paper basket fixed ring is arranged at the waste paper basket; and a fixed ring connecting pipe is arranged at the right side of the paper basket fixed ring. A campus card identifier is arranged at the right side of the lower limb extending groove; a sum display screen is arranged at the front side of the campus card identifier; and an identifier switch is arranged at the right side of the sum display screen. According to the invention, the structure is simple and the operation is convenient. Knowledge points can be searched rapidly by students; the learning efficiency of the student can be improved; and the learning burden can be substantially reduced.