2118results about How to "Improve learning effect" patented technology

The invention provides an augmented-reality-based three-dimensional interactive learning system and an augmented-reality-based three-dimensional interactive learning method. The system comprises an information processing device, a photographic device, a display device and at least one physical teaching aid, wherein the physical teaching aid is provided with identification information; the photographic device is used for performing video acquisition on a real environment after augmented reality application is started; and the information processing device comprises an identification module for identifying the identification information on the physical teaching aid, an orientation calculation module for calculating the spatial orientation information of the physical teaching aid, and a three-dimensional rendering module for acquiring a three-dimensional model corresponding to the identification information, rendering and generating a corresponding virtual object, and placing the virtual object at a corresponding position in a video image for display according to the spatial orientation information of the physical teaching aid. According to the system and the method, the real environment and the virtual object are overlapped in the same scenario in real time, so that a more vivid sensory experience is provided for a user, and meanwhile, a teaching effect is improved by utilizing the instinct of a person for the recognition of a three-dimensional space.

The invention provides a simulation training device for manual arc welding rod-moving operation, and an arc welding rod-moving detection method, which is used for the introduction, improvement and enhancement training of welders in welding rod-moving operation. The device comprises a simulated welding torch, a simulated test plate, an image detection module, a dip-angle sensor, a master control computer and an evaluation system. Through the reasonable choice and layout of sensors and efficient data processing, high-precision detection and evaluation of real-time positions of simulated welding rods operated by the welders are realized. The device can record the arc striking, rod moving, ending and a series of operation of welding students during operation, and allows operation tracks to reappear after the operation is completed so as to improve learning efficiency and learning effects.

The invention provides an automatic segmentation method for an MRI (Magnetic Resonance Imaging) image brain tumor based on a full convolutional network. The method comprises multi-mode MRI image preprocessing of the brain tumor, construction of a full convolutional network model, network training and parameter optimization as well as automatic segmentation of a brain tumor image, specifically, the segmentation of the MRI image brain tumor is converted into a pixel-level semantic annotation problem and differential information emphasizing different modes of MRI, two-dimensional whole slices of four modes FLAIR, T1, T1c and T2 are synthesized into a four-channel input image, the convolutional layer and the pooling layer of the trained convolutional neural network are base feature layers, three convolutional layers equal to a full connection layer are added behind the base feature layers to form a middle layer, the middle layer outputs rough segmentation images corresponding to semantic segmentation types in quantity, and a de-convolutional network is added behind the middle layer and used for interpolating the rough segmentation images to obtain a fine segmentation image having the same size as the original image. The method does not need manual intervention, effectively improves the segmentation precision and efficiency, and shortens the training time.

The invention discloses a log management and fault diagnosis method and device under multiple hosts. The method comprises fault log collection, fault log analysis and fault log correlation analysis. Fault log collection comprises collecting fault logs of all hardware and software in the cluster and storing the logs in a log server uniformly. Fault log analysis comprises filtering the fault logs, extracting log template information and classifying the logs according to log types. Fault log correlation analysis comprises performing fault reason analysis by using log analysis results and combining time windows, clustering the related faults caused by the same fault into a category and trying to find the root of the type of the faults. The method and the device are capable of analyzing the system operation condition effectively, assisting administrators to process system faults and improving the fault type determination accuracy.

The invention provides an Internet teaching method and system for developing teaching activities through the Internet including a wireless communication network, and particularly discloses a method and system capable of recording the Internet teaching process in real time and accurately and efficiently storing the Internet teaching process in a cloud side. The method includes the step of evaluating educational document learning results by students to learn and distinguish learning situations of students of different levels, thereby achieving targeted differential teaching, tutoring, homework doing, exercising and the like and easily improving the learning efficiency and learning effect, especially continuously improving the students at different learning progresses. Furthermore, by means of a data return gateway, a data verification mechanism, a timestamp verification mechanism, an audio and video flow verification mechanism, a data complementary transmitting mechanism, a local fragmentation encrypting, storing and uploading mechanism and other means, recorded data is stored in the cloud side efficiently and accurately.

The invention relates to a classroom teaching recording and video-on-demand method and a system thereof and aims at increasing effects of course review and studying consolidation. During a course recording process, through a time stamp identification, recorded classroom teaching data is segmented; each segment of recorded data corresponds to at least one corresponding course key point; according to the course key point, after each segment of data is recorded or a whole course is recorded, a problem corresponding to each course key point is additionally arranged; further, each course key point and after-class homework or each test question of a classroom test are established corresponding associations, and corresponding relations of the course key points, the segmented identification data, problems and test questions are used to construct a corresponding association database so that if a student has a problem when answering each problem or the test questions after watching a segment of content or all the contents of the online video-on-demand recorded course, he / she can realize high-efficiency course review and the studying consolidation of the recorded course only through the corresponding association of the key points and the segmented identification data and only through reviewing the corresponding segment of recorded data.

The invention provides a Chinese text classification method based on a super-deep convolution neural network structure model. The method comprises the steps of collecting a training corpus of a word vector from the internet, combining a Chinese word segmentation algorithm to conduct word segmentation on the training corpus, and obtaining a word vector model; collecting news of multiple Chinese news websites from the internet, and marking the category of the news as a corpus set for text classification, wherein the corpus set is divided into a training set corpus and a test set corpus; conducting word segmentation on the training set corpus and the test set corpus respectively, and then obtaining the word vectors corresponding to the training set corpus and the test set corpus respectively by utilizing the word vector model; establishing the super-deep convolution neural network structure model; inputting the word vector corresponding to the training set corpus into the super-deep convolution neural network structure model, and conducting training and obtaining a text classification model; inputting the Chinese text which needs to be sorted into the word vector model, obtaining the word vector of the Chinese text which needs to be classified, and then inputting the word vector into the text classification model to complete the Chinese text classification.

The invention discloses a big data based wireless real-time position positioning method. The big data based wireless real-time position positioning method comprises the following steps of gridding a positioned detection area, enabling a plurality of wireless routers which are arranged inside the detection area to transmit wireless signals, enabling a wireless signal receiver to collect wireless signal information of the wireless routers for a plurality of times in every grid of the area after gridding and forming collected wireless signal information into a big database wireless signal fingerprint database; performing signal preprocessing; performing parameter training; training DNN(Deep Neural Networks); performing signal character extraction and character classification based on the trained DNN; performing position estimation based on a HMM (Hidden Markov Model). The big data based wireless real-time position positioning method has the advantages of improving the accuracy of a positioning result under the condition that the real-time positioning speed is not influenced, successfully integrating positioning problems into the big data background and improving the performance of the real-time positioning system by effectively using advantages of the big data.

The invention discloses a real-time expression recognition method based on a multichannel parallel convolutional neural network (MPCNN). The method comprises the steps of: extracting expression data, which contains RGB and Depth images, from a facial-expression data set; carrying out local binarization preprocessing (LBP) and facial-key-point extraction preprocessing on the color images, carrying out gradient preprocessing on the depth images, dividing the preprocessed images into two parts of a training set and a test set, and building the multichannel parallel convolutional neural network; sending the preprocessed images in the training set to the network for training to obtain depth channel, lbp channel and key-point channel recognition models which learn facial-expression outlines, stereoscopic distribution and key-point features; and adopting maximum confidence to fuse classification results of the three recognition models to obtain a final expression recognition model, and building a real-time expression recognition system. The method enhances the robustness of the recognition network, and effectively improves the performance of the real-time expression recognition system.

The invention provides a robot reinforced learning initialization method based on a neural network. The neural network has the same topological structure as a robot working space, and each neuron corresponds to a discrete state of a state space. The method comprises the following steps of: evolving the neural network according to the known partial environmental information till reaching a balance state, wherein at the moment, the output value of each neuron represents maximum cumulative return acquired when the corresponding state follows the optimal strategy; defining the initial value of a Q function as the sum of the immediate return of the current state and the maximum converted cumulative return acquired when the subsequent state follows the optimal strategy; and the mapping the known environmental information into the initial value of the Q function by the neural network. Therefore, the prior knowledge is fused into a robot learning system, and the learning capacity of the robot at the initial stage of reinforced learning is improved; and compared with the conventional Q learning algorithm, the method has the advantages of effectively improving the learning efficiency of the initial stage and increasing the algorithm convergence speed.

The invention discloses a method for detecting classroom attention of a student. The method comprises the following steps: acquiring a classroom scene image; positioning a face and calculating an orientation pose of the face; converting a two-dimensional position of the face in the image into a two-dimensional position thereof in a sitting height datum plane in a classroom, and adding a sitting height priori value of the student to obtain a three-dimensional spatial position of the face in the classroom; through combination of the three-dimensional spatial position of the face and the orientation pose of the face, calculating an attention point of the student on a teaching presentation board. The invention further provides a device for implementing the method. The device comprises a first camera, an identity recognition module, a second camera and a monitoring and analysis module, wherein the first camera is used for acquiring a face image of the incoming student at an entrance of the classroom; the identity recognition module is used for acquiring the identity of the student through the face recognition; the second camera is used for acquiring the classroom scene image; the monitoring and analysis module is used for calculating the attention point of the student on the teaching presentation board. After application of the method and the system, the classroom attention of the student can be accurately monitored to help a teacher to adjust the teaching way in time so as to improve the teaching effect.

The invention discloses an ReLU convolutional neutral network-based image denoising method. The method comprises the following steps of building an ReLU convolutional neutral network model, wherein the ReLU convolutional neutral network model comprises a plurality of convolutional layers and active layers after the convolutional layers, wherein the active layers are ReLU functions; selecting a training set, and setting training parameters of the ReLU convolutional neutral network model; training the ReLU convolutional neutral network model by taking a minimized loss function as a target according to the ReLU convolutional neutral network model and the training parameters of the ReLU convolutional neutral network model to form an image denoising neural network model; and inputting a to-be-processed image to the image denoising neural network model, and outputting a denoised image. According to the ReLU convolutional neutral network-based image denoising method disclosed by the invention, the learning ability of the neural network is greatly enhanced, accurate mapping from noisy images to clean images is established, and real-time denoising can be realized.

The invention is applicable to the field of computer technology and provides a generation method and device for a decision network model of vehicle automatic driving. The method comprises the steps that a sample triad corresponding to each test moment is generated according to vehicle state information collected at each test moment, a preset vehicle movement set and a preset return function, all the sample triads are stored as sample data in a pre-established experience database, and all the sample data is subjected to clustering analysis; training samples are uniformly collected from each cluster obtained after clustering analysis of the experience database according to a preset sampling scale value, and a return accumulated value of each training sample is calculated; and according to all the training samples, the return accumulated value of each training sample and a preset deep learning algorithm, training is performed to obtain the decision network model of vehicle automatic driving. Therefore, the training efficiency of the decision network model and the generalization ability of the decision network model are effectively improved.

The invention discloses an adaptive fault detection method for an airplane rotation actuator driving device based on deep learning. According to the invention, adaptive fault detection is carried out on the airplane rotation actuator driving device based on a sparse Dropout automatic coder and a noise reduction automatic coder and deep learning of Logistic regression, feature self-learning of original data is realized through using the Dropout automatic coder in a first layer and a layered noise reduction automatic coder model in a second layer and a third layer by adopting a multi-layer neural network based deep learning autonomous cognitive method, data features acquired by learning are inputted to a Logistic regression model so as to judge an operating state of the rotation actuator driving device, a threshold is enabled to change along with different inputs and different states of the system through additionally arranging an adaptive threshold fault observer, and a residual error caused by non faults is eliminated. The method disclosed by the invention can be effectively applied to fault diagnosis of the airplane rotation actuator driving device.

The invention provides a video perception-fused multi-task synergetic recognition method and system, and belongs to the technical field of multisource heterogeneous video data processing and recognition. According to the method and system, a biological vision perception mechanism is combined to research feature-synergetic shared semantic descriptions of multisource heterogeneous video data and universal feature descriptions of the multisource heterogeneous video data are obtained; an environment-suitable calculation theory is utilized to establish task-synergetic feature association learning and task prediction mechanism and realize an environment-suitable perception task association prediction mechanism; and long-time dependency is combined to put forward a context-synergetic vision multi-task deep synergetic recognition mode, realize a multi-task deep synergetic recognition model with long-time memory and solve the problem that the video multi-task recognition is bad in generalization, low in robustness and high in calculation complexity. According to the method and system, an intelligent, generalized and mobile video common feature description method and the multi-task deep synergetic recognition model are put forward, so that the development in the field of intelligent information push and personalized control services of smart city multisource heterogeneous video data canbe prompted.

The invention discloses a children cognitive system based on augment reality technology and a cognitive method. The children cognitive system based on the augment reality technology comprises an image information input module, an image information matching module, a drawing model module and a voice recognition module. A set of recognition identifications are developed through interfaces of an ARToolkit augment reality development kit, a Microsoft Speech SDK voice recognition engine, a 3D max modeling tool and the like, and a children cognitive platform performs simple interaction with a virtual scene in a computer through voice recognition. Operations like moving, amplifying and contracting a model are achieved through OpenGL image processing technology and 3D modeling technology. The children cognitive system based on augment reality technology and the cognitive method have the advantages of being short in development cycle, good in maintenance, good in portability and easy to modify. In addition, a user can use the cognitive system based on augment reality technology and the cognitive method to write a literary handbook, good learning effect is achieved, and the cognitive platform which is strong in interaction is provided for a child to use an augment reality application system.

Provided is a student answer information real-time collection and efficient and intelligent correcting system and use method in a teaching process, belonging to a teaching system. A teacher lesson preparation module is used for teachers to compile exercises, input standard answers and scores, and upload data to a server; students upload answers to the server at student clients; an efficient and intelligent correcting module is used for automatically correcting objective questions, preprocessing subjective questions, automatically correcting the questions capable of being intelligently corrected, sending the questions incapable of being intelligently corrected to teachers for correction, and transmitting correction results to the server; the server feeds back the correction results to the student clients. The system and method can realize ''networking, digital and customized'' education, allow students to ''learn anywhere and anytime'', deeply fuse with current education, reach the effect of ''classroom application, regular application, and universal application'', and be easily and extensively copied in areas with poor education resources to realize educational equity and teaching in accordance with the aptitude of students.

The invention discloses a deep feature and transfer learning-based rolling bearing fault diagnosis method under variable working conditions, relates to the technical field of fault diagnosis, and aimsto solve the problem of low state identification accuracy of different fault positions and different performance degradation degrees of a rolling bearing under the variable working conditions. The method comprises the following steps: firstly, carrying out feature extraction on the vibration signal frequency domain amplitude of the rolling bearing by adopting SDAE to obtain vibration signal deepfeatures, and forming a source domain feature sample set and a target domain feature sample set; then, adopting the JGSA to carry out domain adaptation processing on the source domain feature sample and the target domain feature sample, the purpose of reducing distribution offset and subspace transformation difference of feature samples between domains is achieved, and domain offset between different types of feature samples is reduced. And finally, completing rolling bearing multi-state classification under variable working conditions through a K nearest neighbor algorithm. Compared with other methods, the method disclosed by the invention shows better feature extraction capability under the variable working condition of the rolling bearing, the sample feature visualization effect of therolling bearing is optimal, and the fault diagnosis accuracy of the rolling bearing under the variable working condition is high.

The invention discloses an empirical mode decomposition and deep learning hybrid model-based wind speed prediction method and system. The method comprises the following steps of S1, decomposing an original wind speed time sequence according to empirical mode decomposition so as to obtain a plurality of intrinsic mode functions; S2, establishing a training data set and a test data set for each intrinsic mode function; S3, inputting a training sample, in the training data set, of each intrinsic mode function into a stack type coding network to perform training so as to obtain a wind speed prediction sub-model; S4, inputting the test data set into corresponding wind speed prediction sub-models to perform prediction so as to obtain prediction output values of the wind speed prediction sub-models; and S5, performing combination superposition processing on the prediction output values of the wind speed prediction sub-models to obtain a final overall prediction output value. According to the method and the system, the prediction precision and robustness of the prediction models are effectively improved and higher short-term wind speed prediction precision can be achieved.

The present disclosure is directed to a body-worn sensor-based system for evaluating the biomechanics and the motor adaptation characteristics of postural control during a sport activity such as a golf swing. Various embodiments use sensors such as accelerometers, gyroscopes, and magnetometers to measure the three-dimensional motion of ankle and hip joints. In several embodiments, additional sensors attached to other body segments are used to improve the accuracy of the data or detect particular instants during the swing (e.g., top of back swing, instant of the maximum speed of arm, and instant of ball impact). In a golf embodiment, the system combines the measured data in conjunction with a biomechanical model of the human body to: (1) estimate the two-dimensional sway of the golfer's center of mass; (2) quantify and evaluate the golfer's balance via his / her postural compensatory strategy; and (3) provide visual feedback to the golfer for improving dynamic postural control.

The invention relates to infant formula milk powder and a production method thereof. The infant formula milk powder contains nutrition constituents such as protein, fat, carbohydrate, vitamins, mineral elements and the like with content regulated by the national standards on infant foods. Moreover, whey peptides, collagen peptides and casein phosphopeptides are added in the milk powder. Since the three kinds of binding peptides are added, the invention has the advantages that the immunity of cesarean infants is improved and the disease resistance of the infants is enhanced; the digestive function of the infants is improved, the intestinal mucosa proliferation is promoted and the digestive function is improved; and the development of the nervous system is promoted, the nutrition supply to the nervous system is guaranteed and the memory and the learning ability of the infants are improved.

The invention discloses an image de-noising method based on a convolution pair neural network, comprising the following steps: building a convolution pair neural network model, wherein the convolution pair neural network model includes multiple convolution pairs and corresponding activation layers; selecting a training set, and setting the training parameters of the convolution pair neural network model; according to the convolution pair neural network model and the training parameters thereof, training the convolution pair neural network model with the goal of loss function minimizing to form an image de-noising neural network model; and inputting a to-be-processed image to the image de-noising neural network model, and outputting a de-noised image. Through the image de-noising method based on a convolution pair neural network disclosed by the invention, the learning ability of the neural network is enhanced greatly, accurate mapping from noisy images to clean images is established, and real-time de-noising is realized.

The invention discloses a multilayer neural network language model training method and device based on knowledge distillation. The method comprises the steps that firstly, a BERT language model and amulti-layer BILSTM model are constructed to serve as a teacher model and a student model, the constructed BERT language model comprises six layers of transformers, and the multi-layer BILSTM model comprises three layers of BILSTM networks; then, after the text corpus set is preprocessed, the BERT language model is trained to obtain a trained teacher model; and the preprocessed text corpus set is input into a multilayer BILSTM model to train a student model based on a knowledge distillation technology, and different spatial representations are calculated through linear transformation when an embedding layer, a hiding layer and an output layer in a teacher model are learned. Based on the trained student model, the text can be subjected to vector conversion, and then a downstream network is trained to better classify the text. According to the method, the text pre-training efficiency and the accuracy of the text classification task can be effectively improved.

The invention provides a method and system for performing prediction based on a composite machine learning model. The composite machine learning model consists of at least two kinds of sub models. The method comprises: step A, obtaining a prediction data record; step B, generating a plurality of feature subsets of a prediction sample corresponding to a prediction data record based on attribute information of the prediction data record; and step C, providing the plurality of feature subsets of the prediction sample for the sub models included by the composite machine learning model to obtain a prediction result of the composite machine learning model for the prediction sample. In the composite machine learning model, the sub models are formed by training according to a gradient lifting frame. Therefore, many types of sub models can be fused effectively for cooperative work and advantages of all sub models are used fully to obtain a good integrated machine learning result.

The invention relates to a vehicle type identification method based on a support vector machine and used for an earth inductor. The vehicle type identification method includes the following steps: vehicle type waveform data which require to be identified are collected by the earth inductor; a plurality of numeralization features are extracted from waveforms, effective data are screened out, and the features are normalized; multilayer feature selection is performed according to the extracted features, and an optimal feature combination is picked out; a vehicle type classification algorithm based on the clustering support vector machine is built, and parameters in a classification function are optimized by adopting a particle swarm optimization algorithm; a binary tree classification mode is built, classifiers on all classification nodes are trained, and a complete classification decision tree is built; and earth induction waveforms of a vehicle type to be identified are input to obtain identification results of the vehicle type. The vehicle type identification method builds a waveform feature extraction and selection mode, simultaneously adopts the classification algorithm based on the support vector machine and the particle swarm optimization algorithm, greatly improves machine learning efficiency, and enables a machine to identify vehicle types rapidly and accurately.

The invention belongs to the technical field of digital image processing, and particularly relates to a sensitive image identification method based on depth learning. The method mainly comprises following steps of preprocessing sensitive images; dividing all preprocessed sensitive image databases into a training set and a test set, wherein the training set is divided into a train part and a validation part with proportion of 5:1, and using the images of the training set in a depth convolution neural network training, wherein parameters between layers of the convolution neural network are obtained through the training; after the training is finished, using a trained model to initializing a test neural network, wherein the test neural network is the same as a trained grid in structure; and inputting test sensitive images into the initialized depth neural network for identification test, thereby achieving identification of the sensitive images. According to the invention, functions of extraction and classification of features can be finished without manual participation and adjustment, and the reliable sensitive image identification method with high performance is provided.

The invention provides an on-line learning platform system based on the Web. The on-line learning platform system is carried by a Web server and a data base server to enable users to conduct on-line learning activities through the network. The on-line learning platform system comprises a course administrator subsystem, a student subsystem, a teacher subsystem, an academic administrator subsystem, a shopping mall subsystem and a headquarter administrator subsystem. The subsystems are independent with each other and share information. By means of the on-line learning platform system based on the Web, the learning enjoyment of students and interactivity of the students and teachers are increased, and the learning effect is improved.