70results about How to "Helps to learn" patented technology

The invention discloses a self-learning video fire detection method. According to the technical field, the method comprises the steps as follows: a batch of fireworks images are collected, the backgrounds of the images are removed and tags are added, and the images are taken as a seed data set; data samples for a specific application environment are automatically generated using the seed data and a collected video; the data samples are trained through an online learning algorithm to get fire and smoke detectors; a front end sends an acquired suspected fire image to a detection server to judge whether there is a fire; and in the running process, the detection server gets new samples constantly to update the detectors online. Compared with the prior art, the method has the following advantages: the method can be used in all kinds of scenarios, and can adaptively improve the performance and reduce the rate of false report; and suspected area extraction and fire identification are separated, and fire identification can be conducted remotely or even be deployed at the cloud, which makes the whole system more convenient to upgrade and maintain.

The invention discloses a human age estimation method based on self-adaptation sign distribution. According to the method, through a self-adaptation training method combined with the application of an age sign distribution model algorithm, the obtained facial image feature vectors and initial age sign distribution are used as inputs, the KL divergence between the input initial age sign distribution and model prediction sign distribution is minimized by applying the age sign distribution model algorithm, and therefore the prediction age sign distribution can be obtained. Then, the prediction age sign distribution of the same age can be used for studying the age sign distribution of the corresponding age, circulation is carried out until training convergence is achieved or the set largest number of times is reached, and then the final prediction model can be obtained. Facial image features to be estimated are input into the final prediction model, and then the prediction age can be output. The age estimation accuracy of the system can reach the level similar to that of the human.

The invention relates to a zero-sample image classification model based on a repeated attention network, and the model comprises a repeated attention network module which is used for training and obtaining the sequence information of an image region; a generative adversarial network module, used for acquiring visual error information; a visual feature extraction network processing module, used forobtaining a one-dimensional visual feature vector of the image; an attribute semantic conversion network module, used for mapping a low-dimensional attribute semantic vector to a high-dimensional feature vector with the same dimension as the visual feature vector by using two linear activation layers; a vision-attribute semantic connection network, used for realizing fusion of a vision feature vector and an attribute semantic feature vector; and a score classification result and reward output module, used for classifying the types with the labels seen by adopting cross entropy loss, and reward output being used for punishing the unseen data without the labels and punishing the prediction result with the highest possibility of the seen types and the unseen types in the data without the labels. According to the invention, the problem of image category label missing can be effectively solved.

The invention discloses a structure for displaying knowledge network nodes of literature resources. The structure comprises a main literature module, a reference module and a pertinent literature module, wherein the main literature module comprises a mechanism unit, a name unit and a content keyword unit. The main literature module is in linkage with the reference module and the pertinent literature module respectively through the mechanism unit, the name unit and the content keyword unit to obtain pertinent literature and full texts of the mechanism unit, the name unit and the content keyword unit in corresponding databases. The knowledge network nodes are browsing page for providing information and expansion functions of single literatures, comprise bibliography and abstract of the single literatures, and are collecting points of various expansion information of the literature. The expansion information reveals incidence relation between knowledge through methods of conceptual dependency, fact dependency and the like, the purpose of expanding knowledge is achieved, and learning, discovering and obtaining of new knowledge are facilitated.

The invention discloses an electrocardiogram analyzing system based on a gold standard database. The system at least comprises a database subsystem (1) and an application service subsystem (2), wherein the database subsystem (1) at least comprises a gold standard electrocardiogram case database (4) and an index database (5); the application service subsystem (2) is used for receiving electrocardiogram data to be diagnosed, which is derived from an electrocardiogram data source (12) and calculating the characteristics of an electrocardiogram to be diagnosed by waveform analysis, characteristic calculation and index matching; the application service subsystem (2) is used for retrieving the index database (5) and the electrocardiogram case database (4) by utilizing the characteristics of the electrocardiogram to be diagnosed to obtain a gold standard electrocardiogram case set which is matched with the characteristics of the electrocardiogram to be diagnosed; the system also comprises an index configuring subsystem (3); the index configuring subsystem (3) comprises an index editor (22) and an index evaluating tool (23); the system can also comprises a knowledge management subsystem (25); the knowledge management subsystem (25) is used for constructing an diagnostic model for representing a causal relationship of a characteristic code and an diagnostic code by scanning the gold standard case database (4); and the application service subsystem (2) is used for directly converting the characteristics of the electrocardiogram to be diagnosed into a diagnostic result by utilizing the diagnostic model.

The equipment for broadcasting correlative phonetics and text includes memory unit, control unit, broadcast unit and display unit. The memory unit is used to store correlative digital phonetic data and text information data. When some digital phonetic data are selected for broadcast, the control unit picks up the corresponding text information data based on the present broadcast time of the phonetic data and outputs them to the display unit for synchronous display. When some information data corresponding to one broadcast time interval are selected of display, the control unit drives the broadcast unit to broadcast the corresponding phonetic data synchronously.

The invention discloses a direct current motor control teaching experiment system. The direct current motor control teaching experiment system includes an upper computer, a control processor and direct current motors; the upper computer automatically generates codes through MATLAB / Simulink model based design; the codes are downloaded to the control processor; the control processor transmits corresponding control signals to the direct current motors; the direct current motors are connected with detection sensors; the detection sensors return acquired detection signals to the control processor; the upper computer monitors the detection signals; the direct current motors include a first direct current motor and a second direct current motor which are connected with each other through a mechanical coupling transmission device; the mechanical coupling transmission device is connected with a baffle used for fixing the transmission shaft of the mechanical coupling transmission device; and the transmission shaft of the mechanical coupling transmission device is provided with a torque transducer connected with the control processor. The direct current motor control teaching experiment system of the invention has the advantages of simple operation and easiness in establishment and experiment verification. With the direct current motor control teaching experiment system adopted, a plurality of teaching experiment contents such as single-motor control and multi-motor control can be realized.

The invention discloses a multi-modal sentiment classification method based on a heterogeneous fusion network, and belongs to the technical field of opinion mining and sentiment analysis. The method comprises the steps of 1) preprocessing video data; 2) constructing a text feature vector and identifying a text emotion category; 3) constructing a picture feature vector and identifying a picture emotion category; 4) constructing an audio feature vector and identifying an audio emotion category; 5) constructing a multi-modal global feature vector and identifying a multi-modal global emotion category; 6) constructing a multi-modal local feature vector and identifying a multi-modal local emotion category; and 7) adopting a voting strategy to obtain a final sentiment classification result. The heterogeneous fusion network adopts two fusion forms of intra-modal fusion and inter-modal fusion, two fusion angles of macroscopic fusion and microscopic fusion, and two fusion strategies of feature layer fusion and decision-making layer fusion. According to the method, the implied associated information among the multi-modal data can be deeply mined, and mutual complementation and fusion among the multi-modal data are realized, so that the accuracy of multi-modal sentiment classification is improved.

The invention discloses a listening sound spectrograph which comprises a musical sound acquiring device, a memory, a feature extraction device, a musical score generating device, a musical score displaying device, a musical score optimizing device and a musical score playing device. The output end of the musical sound acquiring device is connected with the input interface of the memory. The feature extraction device is bi-directionally communicatively connected to the memory. The musical score generating device bi-directionally communicates with the memory. The musical score optimizing device bi-directionally communicates with the memory. The input end of the musical score displaying device is connected to the output interface of the memory. The input end of the musical score displaying device is connected with the output interface of the memory. The listening sound spectrograph realizes automatic music identification, recording, and proofreading on a mobile terminal, helps to learn a basic music theory, the promotion and popularization of music art.

The invention discloses an intelligent hand pen capable of rectifying disorder written characters and wrongly written characters, which comprises a pen holder with a cavity, a pen cap, a pen body, a memory and a central processor, wherein the pen cap is arranged on the lower end of the pen holder, the pen body, the memory and the central processor are arranged in the pen holder, a light sensor is arranged on a pen point of the pen body, a loudspeaker is arranged on the pen holder, the central processor is respectively connected with the loudspeaker, the light sensor and the memory, and comprises an information receiving module, a comparison module, a timing module and a control module, the memory comprises a first and a second memories, Chinese characters, correct stroke sorting, words comprising each Chinese character and idioms codes are arranged in the first memory, the second memory and the light sensor are connected through the central processor, and the first and the second memories are also connected with the comparison module of the central processor each other. The intelligent hand pen capable of rectifying the disorder written characters and the wrongly written characters has the alarm function of the disorder written characters and the wrongly written characters, and is contribute for users to rectify the disorder written characters and the wrongly written characters in time.

The invention discloses a face age estimation method based on a sparse undirected probabilistic graphical model and belongs to the technical fields of machine learning and mode identification.According to the face age estimation method, an age distribution prediction model is innovatively constructed through an undirected probabilistic graph, and in addition, proper sparsity regular terms are added to an optimization training target of the model so as to constrain model parameters.The invention further discloses a face age estimation system based on the sparse undirected probabilistic graphical model.Compared with the prior art, the face age estimation method and system have the advantages that more abundant information can be learned out of complex image features so as to predict age distribution, and in addition, more compact encoding can be conducted on the information through term vectors; the experience of image sparsity is utilized, and the sparsity regular terms are added to constrain the model parameters, so that model obtained through learning has better generalization.

The invention relates to a hybrid digital-analogue simulation system for a static var generator and a simulation method for the hybrid digital-analogue simulation system. The simulation system is used for technical skill training, and comprises a low-voltage physical model, a low-voltage physical model control system and a digital simulator, wherein the low-voltage physical model is controlled by the low-voltage physical model control system when independently running, and is controlled by the digital simulator during hybrid digital-analogue simulation; a power module in the digital simulator can be controlled by the low-voltage physical model control system. According to the system and the method, the low-voltage physical model can be locally / remotely operated, real-time online simulation researches can also be made, diversified teaching modes can be flexibly adopted, and good demonstration teaching effects can be achieved.

The invention discloses a desk, and particularly discloses an adjustable desk. The adjustable desk is characterized in that a desktop is arranged at the top part of a desk body, the desktop is internally provided with a push-pull board, supporting rods are arranged at the bottom part of the push-pull board, plug-in buttons are arranged on the desk body, telescopic desk legs are arranged at the lower part of the desk body, and clamping grooves are arranged in the telescopic desk legs. The adjustable desk disclosed by the invention has the beneficial effects that the height and the width of the desk can be adjusted according to needs, students in different sizes can use the desk due to the effective assistance of the adjustable desk, and the adjustable desk is beneficial for the study of the students.

The invention belongs to the technical field of computers, and particularly relates to an action classifying method for small sample videos. Aiming at solving the problem that the types of a trainingset and a test set are intersected in the existing setting of small sample video action classification, the invention provides a small sample video recognition new mode based on an intelligent human body, generates a large number of virtual videos with the same action in a 3D intelligent virtual human body and virtual environment interaction mode, and provides a training sample for a deep neural network. In addition, the invention also provides a data enhancement method based on video segment replacement, capable of expanding a limited data set by means of a method of replacing a certain segment in an original video with video segments with similar semantics. Experiments show that the method can play a great role in promoting small sample video action recognition, and has good robustness and strong algorithm transportability.

The invention provides a study aid for British and American literature education and teaching, and relates to a teaching device. A groove is formed in the front side surface of a comparison study plate; a word group collocation comparison study device movably connected with the comparison study plate is arranged on the inner bottom surface of the groove; an opening is formed in one side of the groove formed in the comparison study plate; and a shielding plate and a writing plate which are glidingly matched with the groove are sequentially arranged in the opening position of the groove from inside to outside. By using the study aid, the comprehensive and diversified display of the collocation of one verb and a plurality of preposition can be realized; the study aid helps students remember the verbs and all prepositions capable of being collocated with the verbs; in addition, practical teaching on English tenses can also be performed; the systematic and regular display of relevant changesentence patterns of 16 tenses of English can be realized; the study aid is favorable for integrally learning all tenses of English by the students; the interactive study between teachers and students and between students is realized; and the English study interestingness is greatly improved.

The invention discloses a character detection method and device based on single characters and inter-character connection relationship prediction. The method comprises the following steps of: traininga neural network model; performing feature extraction through the neural network model to obtain a feature map which comprises a character position score map Text-map and an inter-character connection relationship score map Link-map; carrying out post-processing on the Text-map and the Link-map; and calculating the minimum bounding rectangle of each connected domain to realize detection of the position of a character from a picture. The convolution kernel of the prediction model based on the character level can be forced to pay more attention to the character-level features, the character-level features learned by the model can be effectively associated at the post-processing branch, long text detection is achieved, and missing detection of characters is avoided.

The invention discloses a robot with a toy function. The robot comprises a robot body which comprises a chassis, a robot shell arranged on the chassis, robot arms and a robot head. A bin gate is arranged on the obverse side of the robot shell and connected with a containing bin, and driving motors connected with a robot control unit are arranged on the two sides of the containing bin. Driving gears are arranged on the driving motors and connected with racks arranged on the two sides of the bin gate. A wireless charging transmitting module is arranged at the bottom of the containing bin. The robot is novel in design, simple and reasonable in structure and convenient to use, can serve as a toy to listen to music, can also be used for early education learning, and can monitor the surroundingenvironment during using to real-timely transmit shot images of the surrounding environment to an electronic display screen on the robot body to be displayed, and remote observation and monitoring arefacilitated.

The invention provides a teaching robot and a teaching machine system, and relates to the technical field of teaching based on robots. The teaching robot comprises a processor, a sensing component, abalance motor, a driving motor and a motion component, the processor is connected with the sensing component, the balance motor and the driving motor, the sensing component can be used for acquiring an image, the balance motor receives a balance control instruction generated by the processor, the motion component is in transmission connection with the driving motor, and the driving motor receivesa target object determined by the processor based on the image and a motion control instruction generated by position information of the target object and drives the motion component to move, so thatthe teaching robot traces the target object. The teaching robot and the teaching machine system are simple in structure and easy to implement, the image can be automatically acquired, and the target object is traced according to the determined target object, so that the system helps people to learn, develop and practice a visual system of the robot.

The invention relates to a table lamp, in particular to a rotatable portable eye-caring LED table lamp. The table lamp comprises a first lamp post, a second lamp post, a third lamp post and a fixed base, a limit nut is arranged at the bottom of the fixed base and is in through connection with the fixed base through a fixing nut, and a first fixed block is arranged on the top of the fixed base androtatably connected with the fixed base through a horizontal rotating shaft; the third lamp post is arranged on the top of the first fixed block and rotatably connected with the first fixed block through a first perpendicular rotating shaft, a charging interface is formed in one side of the third lamp post, and a luminance sensor, a control switch, a Bluetooth switch and a microphone are arrangedon the base surface of the third lamp post; a controller and an audio module are arranged in the third lamp post, and the second lamp post is arranged at the top end of the third lamp post. The LED lamp is simple in structure, the illuminating angle can be conveniently adjusted, and certain popularization effect is achieved.

The invention discloses a T cell receptor sequence classification method based on a semi-supervised learning framework, and the method comprises the steps: selecting a CDR3 beta region as input data,and carrying out the feature coding of T cell receptor data; according to the obtained data, selecting supervised learning algorithms of a support vector machine, a random forest and a decision tree to respectively construct initial classifiers C1, C2 and C3; training the initial classifiers C1, C2 and C3 to obtain expanded new training sets, repeatedly sampling the generated training sets to obtain three marked training sets, generating a classifier from each new training set, and iteratively updating the classifiers; and after the training is finished, using the three classifiers C1, C2 andC3 as a classifier integration through a voting mechanism. The method is suitable for the situation that T cell receptor sequence data is difficult to obtain, and the performance is remarkably superior to that of an existing method.

Disclosed is a menu broadcasting system applied in cooker hood. The system is characterized by comprising: a central processing unit; a microphone, a speaker, and a wireless communication module, all of which are connected to the central processing unit. The central processing unit recognizes voice signals received by the microphone of the central processor, then receive and send required menus via the wireless communication module, broadcast menus via the speaker whereafter. In operation, the system first receive inputs of user voice via the microphone, then pass the inputs through the central processing unit, the wireless communication module uploads the inputs to a network server which then converts the voice inputs to text inputs, thereby retrieving menus, sending menus retrieved to the central processing unit, and broadcasting the menus to users via the speaker at the end. Users can use the system to retrieve menus conveniently, quickly, and intelligently; the system helps users who can't cook learn cooking a variety of cuisine. The system greatly improves cooking quality.

The invention discloses a multifunctional music teaching device. The device comprises a bottom plate, supporting legs are arranged on the lower surface of the bottom plate, a musical instrument box isarranged on the right side of the upper surface of the bottom plate, and a partition plate is arranged in the musical instrument box; the partition plate divides the interior of the musical instrument box into a first storage box and a second storage box, the opened end of the musical instrument box is hinged to a cover plate through a hinge, and an electric telescopic rod switch and a music player switch are arranged on the right side of the upper surface of the bottom plate; a music stand is arranged at the position, close to the electric telescopic rod switch, of the right side of the upper surface of the bottom plate; a music player is arranged at the position, close to the music stand, of the right side of the upper surface of the bottom plate. The multifunctional music teaching device is convenient to use, diversified in function and convenient to move, musical instruments can be stored in a classified mode, the musical instruments can be well protected and do not need to be carried by music teachers many times, convenience is brought to the music teachers, the height of a supporting plate can be adjusted, and teaching is facilitated; a frame type baffle and elastic bandagecan well fix a music score, and the music player can help students to learn.

The invention discloses an alarm clock for primary school students with the function of reading, which comprises a clock face, a back cover, supporting legs and an alarm clock body. One side of the alarm clock body is a clock face and the other side is a back cover. Hour hand, minute hand and second hand, speakers are fixed on both sides of the clock face, banknote inspection lamps and strong lights are respectively installed at the upper and lower ends of the clock face, the speaker, currency inspection lamp and strong light are all located on the alarm clock body, and the top of the alarm clock body A miniature fan is installed, and LED colorful lights are installed on both sides of the miniature fan. There are three supporting feet at the bottom of the alarm clock body, and a capacitive touch screen is embedded above the back cover. There is a storage battery embedded under the adjustment button and the light switch group. Through the setting of the capacitive touch screen, the alarm bell can be adjusted to the content of the text, which can not only help wake up the pupils, but also increase the impression of the text for the pupils. The design of the invention is reasonable and suitable for Promotional use.

A swimming suit with buoyance is composed of an upper wear, a waist band and a swimming pants, all of which can be easily assembled together or dissembled from each other and have the floating blocks with the buoyance ratio of (20-25%):(45-55%):(25-35%).

The invention relates to a teaching aid, in particular to a TRIZ (theory of inventive problem solving) teaching playing card. Core knowledge points, technical terms, theoretical systems and cases are uniformly arranged in the playing card, keywords, detailed contents and engineering cases are respectively arranged on front surfaces and back surfaces of playing card units, TRIZ knowledge is integrated into the playing card popular for people and is learned when the playing card is played, and learning interestingness and learning efficiency are improved advantageously.

The invention discloses a multifunctional wickerwork floor lamp. The multifunctional wickerwork floor lamp comprises a base, a supporting column, a lampshade and a light source and further comprises a supporting ball, light sensing lamps, a loudspeaker, a display screen and an intelligent module. The lampshade is of a hollow conical structure formed by nail joint of multiple branches. A keel framework of a grid structure is in nail joint to the inner side of the lampshade. A lamp base is arranged on the top face of the bottom of the lampshade. The light source is arranged in the lamp base. The light sensing lamps are embedded in the surface of the outer side of the lampshade. The hollow supporting ball is in nail joint to the lower end of the lampshade. The hollow supporting column is connected to the bottom of the supporting ball in an inserted manner. The loudspeaker and the display screen are arranged on the front side face of the supporting column. The bottom end of the supporting column is inserted in the base in a connected manner. Touch control buttons are arranged on the front side face of the supporting ball. The intelligent module is fixedly arranged in a cavity of the supporting ball and comprises a power module, a control module, a U disk read-write module and an audio module. The control module comprises a timer, a CPU and a relay. The multifunctional wickerwork floor lamp has the multiple functions of low carbon, environment-friendliness, fatigue relieving and teaching intelligence benefiting.

The invention discloses a medical data processing system and method. The medical data processing system comprises a data receiving module, a text data processing module, an image data processing module, a text database, an image data server, a client, a data correction module and a data transmission module. According to the invention, detection data from a detection instrument is processed throughthe text data processing module and the image data processing module; the detection data is matched and associated with patient application information from a report generation system, the detectiondata is comprehensively judged and corrected through a data correction module, and finally the corrected detection data is automatically sent to the report generation system; therefore, the manual image detection time is greatly shortened, and the working efficiency is greatly improved; and the problems of misinformation, missing information and the like can be effectively avoided, and the reliability of an image interpretation result is further improved.

The invention provides a process equipment and control engineering comprehensive experiment system. The process equipment and control engineering comprehensive experiment system includes a boiler, a compressor, an expander, a heat exchanger, a condenser, a centrifugal pump and a control box, wherein during the operating process, a cycle fluid and a condenser heat exchange medium take part in the process; the boiler heats the liquid cycle fluid into the steam state; the compressor pressurizes the gaseous state cycle fluid; the expander enables the cycle fluid to do work by expansion; the heat exchanger and the condenser perform condensation cooling on the gaseous state cycle fluid; the centrifugal pump pressurizes the cycle fluid, and sends the cycle fluid to the boiler; and the control boxis used to monitor and control the whole experiment equipment. The process equipment and control engineering comprehensive experiment system is miniature, and various devices in the experiment systemare put in a laboratory. The process equipment and control engineering comprehensive experiment system does not need providing extra heat sources for evaporation of cycle fluid, and is convenient foroperation of the experiment process by electrically heating the cycle fluid. The process equipment and control engineering comprehensive experiment system can achieve the aim of cycling fluid steam at the entrance of the expander generating the specified state parameters in a mode of combining the boiler and the compressor.

The invention discloses a multipurpose lamp which comprises an LED light-emitting module, a holding and lifting part, a lamp base, a suspension part, a power supply device and a stored electricity supplying device. The multipurpose lamp is reasonable in design. Multiple using states can be realized due to the arrangement of the holding and lifting part, the suspension part and the lamp base, and the limitation that a table lamp can only be used in a room is broken. The holding and lifting part enables the suspending illumination function to be realized and is more suitable for outdoor use. The lamp can be hung on a wall easily through the suspension part to meet different personalized requirements. The illumination angle can be adjusted freely through a universal connector, and using is convenient. Furthermore, both the power supply device and the stored electricity supplying device are available, an LED can be supplied with power anytime anywhere through the power supply device or the stored electricity supplying device even in outdoor places where power supplies are hard to find, using is convenient and safe, movability is high, carrying is convenient, people can study or entertain themselves anytime anywhere, and the application range is wide.

The invention discloses a buoyancy demonstration teaching aid for a physical experiment, wherein the teaching aid comprises an operation platform, and the top of the operation platform is provided with two containers. Under the action of springs II, rubber blocks are tightly attached to the side walls of stand columns to fixe sliding sleeves, the end part of each sliding sleeve and the initial liquid surfaces of a liquid medium I and a liquid medium II are positioned at the same height; pull rings are pulled, when limiting pins are separated from limiting grooves, under the action of the gravity of floating blocks, a rotary disc rotates, moving blocks move in arc-shaped grooves, and the moving blocks move towards the two sides of a transverse plate I; the two floating blocks respectively fall into two containers, and the floating blocks float on the liquid surfaces of the liquid medium I and the liquid medium II; by observing the rising heights of the liquid medium I and the liquid medium II, the inference that the water discharge amounts of an object in the liquids with different densities are different when the buoyancy borne by the object is fixed can be verified more visually and vividly, and teaching of teachers and learning of students on the buoyancy principle are facilitated.