112 results about "Training Supports" patented technology

Solving a quadratic programming problem involved in training support vector machines by sweeping through a set of training examples, solving small sub-problems of the quadratic programming problem. Each of these sub-problems has an analytic solution, which is faster that the numerical quadratic programming solutions used in the prior art. In one embodiment, training examples with non-optimal Lagrange multipliers are adjusted, one at a time, until all are optimal (e.g. until all examples fulfill the Kuhn-Tucker conditions). In another embodiment, training examples with non-optimal Lagrange multipliers are paired and then adjusted, until all are optimal.

The invention discloses a social network garbage user detection method integrated with multiple algorithms, and the method comprises the steps: obtaining the user data from a social network in a mode of web crawler, and extracting the corresponding features through analyzing the behaviors of users to form a feature vector; carrying out the clustering of users in the social network through employing a clustering algorithm combining K-Means and DBSCAN; selecting the data nearby a cluster boundary and a data training support vector machine (SVM) classifier nearby a clustering center from the clustering result at the previous step, and obtaining a classifier model; and finally detecting the garbage users in the social networks through the SVM classifier model obtained through training. The method reduces the cost of manual marking of data, improves the detection accuracy, and is easy to implement.

A training support system (S) allows an exerciser to practice training exercises safely and effectively with high motivation. The training support system (S) comprises a display means (4) which conducts display relating to training for the exerciser who uses a training machine (1) for muscle training, and a controller (3). The controller (3) allows dynamic display of reference motion and training motion on the display means (4) and variation of the display on the display means (4) according to the difference between the reference motion and the training motion.

The invention discloses a cigarette smoke detection method based on computer vision. The method comprises the two stages of classifier generation and cigarette smoke detection. The classifier generation stage comprises the steps of receiving sample video information, extracting a suspected cigarette smoke area through multichannel background subtraction, extracting motion features of the suspected area and combining the extracted motion features into a feature vector training support vector machine. The cigarette smoke detection stage comprises the steps of receiving the video information to be detected, extracting dynamic features of the suspected cigarette smoke area and combining the dynamic features into a feature vector input classifier through the same method used in the classifier generation stage, and judging whether cigarette smoke exists in the suspected area. By means of the cigarette smoke detection method based on computer vision, large-range cigarette smoke detection in a simple scene can be achieved, and smoke control work in a public place is facilitated.

A customer support system for improving the skill of a person engaged in a business with reduced cost for technical training support and enhancing practical training. The customer support system includes a computer system for introducing training contents from a service providing side to a customer side through a computer network; an instructor training facility for training an instructor based on training contents; and a remote technique training support device for the instructor, for supporting technical training on the customer side, from the service providing side, through the computer network.

A training support system (S) allows an exerciser to practice training exercises safely and effectively with high motivation. The training support system (S) comprises a display means (4) which conducts display relating to training for the exerciser who uses a training machine (1) for muscle training, and a controller (3). The controller (3) allows dynamic display of reference motion and training motion on the display means (4) and variation of the display on the display means (4) according to the difference between the reference motion and the training motion.

The invention belongs to the technical field of SAR (synthetic aperture radar) image change detection, and discloses an SAR image change detection method based on a support vector machine and a discriminative random field. The SAR image change detection method based on the support vector machine and the discriminative random field includes the steps: normalizing gray values of two original time phase images, and extracting corresponding gray characteristic differences and textural characteristic differences in the processed images; forming difference characteristic vectors; extracting boundary strength of each pixel in a difference image by the aid of weighted average ratio operators; selecting training samples in the difference image, and expressing the training samples by the aid of the corresponding difference characteristic vectors to obtain initial category labels of testing samples and posterior probabilities of the category labels of the testing samples by the aid of the training support vector machine; obtaining initial support vector machine-discriminative random field models; updating the support vector machine-discriminative random field models to obtain final category labels and change detection results of the corresponding testing samples.

The invention discloses a method for monitoring and diagnosing a flat wheel state of a rail vehicle based on VMD. The method comprises the following steps of acquiring a vibration signal of a flat wheel through a sensor which is mounted in a rail; performing VMD variation modal decomposition on the vibration signal and obtaining a plurality of modals in different frequencies; constructing an envelope spectrum entropy characteristic for the plurality of modals; selecting an envelope spectrum entropy value with high consistency with the vibration signal from the envelope spectrum entropy characteristic; and inputting a plurality of selected envelope spectrum entropy values into a training support vector machine for performing identification, thereby identifying the working condition of the flat wheel. The method for monitoring and diagnosing the flat wheel state of the rail vehicle based on VMD has an advantage of high flat wheel working condition identification accuracy.

The invention discloses a multi-model mine roof safety early warning model based on decision fusion. Real-time pressure monitoring data are collected through sensors, and an average value is selected as a training support pressure monitoring datum X1 to be used for establishing a support pressure submodel; stress distribution information of a coal bed or a rock stratum is collected through a leading pressure sensor to obtain a training leading pressure detection datum X2, and a leading pressure submodel is established; a separation layer indicator collects the separation layer situation of rock in a roof rock pillar near a working face and conducts processing to obtain a training roof separation layer monitoring datum X3, and a roof separation layer submodel is established; the data in the submodels are reconstructed to form corresponding prediction values, namely, Y1, Y2 and Y3, normalized decision fusion is conducted on the prediction values in a statistical recognition mode, and fused information is obtained. The multi-model mine roof safety early warning model is high in prediction and generalization capacity.

A support vector machine (110) is used to predict who, among a population of patients with diabetes mellitus, will develop proteinuria which is in indicator of diabetic nephropathy. The support vector machine (110) is trained using test results of the patients from blood biochemistry and haemotology tests. The training and testing of the support vector machine (110) used data in which the entire patient population did not exhibit signs of proteinuria at a predetermined time period and three months later, and some of the patient population had proteinuria six months from the predetermined time period. The support vector machine (110) is used to predict who, among patients with diabetes mellitus using lest results from a predetermined time period and three months later, will develop proteinuria at six months from the predetermined time period. The input data to the support vector machine (110) included different parameters of test results at a predetermined time and three months later.

A barbell / dumbbell training support device includes a base with distal ends, a cross support laterally extending equidistantly from each distal end of the base for a predetermined distance, the cross supports being configured to prevent tipping of the support device, an elongated tray configured to support barbells and dumbbells, and connection elements configured to interconnect the base with the elongated tray. Ends of each cross support may have attached thereto pads, covers, or elevated attachments made from flexible and resilient material. The elongated tray is a generally elongated rectangle having a surface and lengthwise sides with ends and widthwise sides with ends. Raised edges extend from each of the corresponding lengthwise and widthwise sides of the tray.

The invention discloses a man-machine gesture interactive system based on an electronic whiteboard. The system comprises a camera for acquiring gesture images, a gesture acquisition module for preprocessing the gesture images, a gesture identification module for extracting image features of the preprocessed gesture images and classifying the image features via a data sample training support vector machine so as to identify the gestures, and an instruction conversion module for converting the identified gestures into electronic whiteboard instructions. In the system, a configurable frame for interaction between the gesture and the electronic whiteboard instruction is provided, the frame is used for mutually mapping the gestures with the electronic whiteboard instructions, and the frame is easy to configure. Simultaneously, with the system, teachers and students need not to stand between the electronic whiteboard and a projector, so that shadow on the electronic whiteboard can be avoided, several people can use the electronic whiteboard at the same time, teaching interactivity is enhanced, and participation of the teachers and the students is facilitated.

The invention provides a fraud recognition method combining with audio analysis and video analysis, a device and a storage medium. The method comprises the following steps: cutting an audio and videosample to obtain audio and video fragments, and distributing a fraud label for each audio and video fragment; decoding and preprocessing each audio and video fragment to obtain an audio fragment and avideo fragment of each audio and video fragment; extracting voice features and expression features from each audio fragment and each video fragment, respectively; combining the voice features of eachaudio fragment and the expression features of each video fragment with a fraud labeling training support vector machine, acquiring a voice analysis model and an expression analysis model; collectingaudio and video data of an object to be recognized; extracting the voice features and the expression features of the audio and video data; inputting the voice features and the expression featrues intothe voice analysis model and the expression analysis model respectively, and outputting the fraud probabilities P1 and P2 of the object to be recognized; and carrying out weighted calculation on P1 and P2, thus obtaining the fraud recognition result of the object to be recognized.

An application that operates on a user terminal records an operation performed by a user, in an operation history data. A training support apparatus stores the operation history data in an operation history DB, and analyzes the operation history data to generate a use record DB. The training support apparatus generates a customized curriculum and advice data based on the use record DB, to provide efficient training according to a knowledge level and a usage pattern of the user.

A method for selling a product and all phases of service and support to a customer for optimal use of the product includes customizing a product for a surgical allograft procedure to be performed and providing the customized product and services as an integrated turnkey package to a medical provider. The products include bone and soft tissue allografts for all types of orthopedic reconstructive procedures and a loan of the associated instrumentation. The service portion of the package includes training support on use of the instruments, training of operating room personnel, technical and customer support both before the performance of the procedure and on-site support. After completion of the surgical procedure, the service provider arranges for the return of the loaned instrumentation.

The invention relates to medical tools, particularly to a novel upper roller transmission and balance device. The novel upper roller transmission and balance device comprises a training support mechanism, a storage mechanism, a support rod, a foot placing mechanism and a protecting mechanism. The novel upper roller transmission and balance device can support the casted leg of a patient to save labor for the patient during walking and help the patient take assisted leg exercise at rest, thereby shortening the curing time for the patient; the height of the novel upper roller transmission and balance device is convenient to adjust to facilitate application by patients of different height and meanwhile to help the patient to take upper limb strength training; meanwhile, the novel upper roller transmission and balance device can store objects carried by the patient to gain convenience; the distance between a handheld connecting sleeve and an axillary support cross bar is adjustable to bring convenience to patients with different arm lengths; the upper end of the storage mechanism is connected with the training support mechanism, the lower end of the storage mechanism is connected with the support rod, the lower end of the support rod is provided with the protecting mechanism, and the foot placing mechanism is mounted on the support rod.

The invention discloses a granular support vector machine ensemble-based protein ligand binding site prediction method. The method comprises the steps of 1, performing feature extraction according to evolution information and a secondary structure of a protein sequence, representing amino acid residues in the sequence in an eigenvector form, and constructing a training sample set by the residues (sites) as units; 2, performing sampling on the training sample set by utilizing a granular computation thought to generate a plurality of sub-training sample sets; 3, training support vector machine (SVM) models on the sub-training sample sets respectively, wherein multiple SVMs form an SVM ensemble; 4, performing integration on the models in the SVM ensemble by adopting an AdaBoost algorithm to obtain an integrated SVM model; and 5, for a given query sequence, generating eigenvectors corresponding to all residues in the sequence by using the same feature extraction method. For each residue sample, the integrated SVM model is used for performing prediction to generate an original prediction result, and then the original result is processed by utilizing a simple post-processing technology to generate a final prediction result. The method is high in prediction precision and good in generalization capability.

The invention relates to a method for judging image memorability based on saliency entropy and object bank feature. A research proves that the image memorability can be influenced by an object contained in an image and the dispersing degree of the image to human visual attraction, so that the object bank feature is used for expressing the object contained in the image and the visual saliency entropy of the image is used for expressing the dispersing degree of the image to the human visual attraction. The two models have excellent expression for the image memorability, so that the two features are combined for expressing a piece of image, and a memorability value of the image with unknown memorability value is forecast through a training supporting regression vector machine model. The method provided by the invention belongs to the field of computer image processing. According to the technical scheme, the image memorability can be judged; the method can be applied to the industries such as advertising industry and news editing; a suitable image can be selected by a practitioner; and the method has a high commercial value.

The invention provides a novel and robust video smoke detection method under a scene with thin smoke. The method is manly composed of three stages, namely a preprocessing stage, a feature extraction stage and an image classification stage. At the preprocessing stage, a background differential algorithm is used to extract a motion foreground region of video frames; then an HSV color space is adopted to act on the motion foreground region to recognize smoke pixels; then a local extreme co-occurrence pattern (LECoP) is used to calculate texture features, and smoke energy analysis is used to calculate energy features; and last, a feature vector training support vector machine (SVM) is used for smoke recognition. It can be seen from an experiment result that the method can effectively detect smoke.

The invention relates to a side somersault training device, which comprises a springboard and a side somersault training support frame, wherein the side somersault training support frame comprises a base framework, upright posts, ascending and descending rods, an upper transverse rod and a lower transverse rod; the upright posts are vertically arranged at the two sides; the ascending and descending rods are arranged in the upright posts; the upper transverse rod and the lower transverse rod are connected with two ends of the ascending and descending rods; two isosceles trapezoid frameworks with the intersected top ends are fixedly connected to the inner side of the upper transverse rods; one end extends to form a picking rod crank arm; the lower transverse rod is rotatably connected with a double-head crank connecting rod in a sleeving way; one end of the double-head crank connecting rod is in contact with a touch switch of the springboard; the other end of the double-head crank connecting rod is a clamp hook rotationally buckled with the lower transverse rod of an isosceles trapezoid framework; the lower transverse rod of the other isosceles trapezoid frame is provided with a spiral spring connected with a transverse post arranged in the middle of the upright posts. The side somersault training device has the advantages that the structure is simple; the moving tracks are precise; the action is specified; the use is safe and reliable; the action study time is reduced; the physical consumption of teachers is reduced; the action conditions of students are watched in a concentrated way; the training is guided; the teaching effect and quality are improved.

The invention relates to the field of voice signal processing, in particular to a data labeling method and device based on self-learning algorithm. The method comprises a speech recognition step, a text comparison step, a natural language processing algorithm evaluation step, a natural language processing algorithm prediction step, a data labeling step, a quality inspection step and a self-learning step. The text comparison step is used for comparing a plurality of recognition texts, labeling difference parts of texts and performing sentence breaking processing. The data labeling step is usedfor performing data labeling on an optimal pre-labeled text for a plurality of times by referring to an original recognition text and a prediction text of the difference parts, so as to form a plurality of groups of data labeling texts. The self-learning step is used for inputting the optimal labeled text and a corresponding audio signal into a speech recognition engine, wherein the speech recognition engine is iteratively trained based on the self-learning algorithm. According to the labeling method and device, the data labeling time is greatly saved, the data labeling quality and the data labeling efficiency are effectively improved, the training support is provided for various artificial intelligence products, and the production effect of intelligent products is improved.

The invention discloses a tank element simulation training system and method. A console computer and a passenger simulation training operation computer are connected to an element training supportingnetwork through a switch; an adapter computer is connected with a simulation conductor task terminal through a small internet controller training simulator and a radio station training simulator and connected to an element guiding control network; a desktop type blue tank simulator is connected to the element training supporting network through the switch; the system adopts a distributed system architecture, the network environment used for information transmission and control is composed of a guiding control network, a training supporting network and a channel simulation network, the trainingsupporting network is mainly used for simulation data interaction, the guiding control network mainly transmits guiding adjusting control information, the channel simulation network completes networksimulation charging communication voice and data transmission under the tactical condition, the red-blue combat simulation training function is achieved, convenience is brought to networking, and extension is facilitated.

The invention discloses a dual-core embedded platform-based system for online training a support vector machine soft-sensing model, which mainly comprises a microcontroller, a memory, a human-computer interaction interface, a communication interface, a signal input and output interface and the like. An embedded platform-based method for online training the support vector machine soft-sensing model of the invention can maximally meet the requirements on real-time performance and intelligence because related parameters can be automatically set according to data characteristics and manual parameter setting is simultaneously supported. The system has a plurality of functions of being portable, supporting dynamic display and intelligent analysis and the like, realizes complex operations through the simple and direct-viewing human-computer interaction interface, overcomes the defects of difficulty in the realization of miniaturization, difficulty in the determination of related training parameters, the complex operations and the like, particularly the successful transplantation of the online training of a support vector machine algorithm, and provides a scheme which is low in cost, portability and high in the real-time performance for soft-sensing in complex processes in petrochemical technology and the like.

The invention provides an image blind detection method based on secondary seam clipping features and belongs to the field of blind image forensics image processing. The method includes: (1) conducting additional seam clipping operation on an image to obtain an additional seam clipping image; (2) conducting feature extraction which includes calculating the similarity feature, the energy ratio feature and the stitch distance difference feature between the image and the corresponding additional seam clipping image; (3) detecting seam clipping through a training support vector machine by utilizing the similarity feature, the energy ratio feature and the stitch distance difference feature which are extracted. The method is low in feature dimension, has lower calculation complexity and meanwhile has higher detection rate.

A computer-readable recording medium having recorded therein agent training program in which when each agent has been performing an answer, agent identifier, and enquiry contents and contents of the answer are stored in a storage device. An enquiry pattern is extracted from enquiry contents. Respective agents' answers are evaluated, for each enquiry pattern, and by a model contents of the answer extraction section, contents of the answer of an agent with highest evaluation is extracted as model contents of the answer. The contents of the answer of a trainee agent and model contents of the answer are divided into predetermined phases, and corresponding phase of both are mutually compared to calculate ineptitude-degrees in respective phases of trainee agent. By specifying from the model contents of the answer, a range of contents of the answer containing therein at least one of the phases where the ineptitude-degree is high, and lasting predetermined time, and generating teaching material with these contents, teaching material specialized for the inept answers of respective agents is provided, and agent training support performed.

Disclosed is a method for early termination in training support vector machines. A support vector machine is iteratively trained based on training examples using an objective function having primal and dual formulations. At each iteration, a termination threshold is calculated based on the current SVM solution. The termination threshold increases with the number of training examples. The termination threshold can be calculated based on the observed variance of the loss for the current SVM solution. The termination threshold is compared to a duality gap between primal and dual formulations at the current SVM solution. When the duality gap is less than the termination threshold, the training is terminated.

The invention discloses a quick sorting method for movie types based on posters and plot summaries. The method comprises: determining a type assembly a movie belongs to, establishing poster training sets and plot summary training sets of various kinds of movies; extracting characteristics of a poster of a to-be-tested movie, using the obtained characteristics of each poster and a label training support vector machine corresponding to the poster to obtain a sorting model of posters; extracting characteristics of a text of the plot summary of the to-be-tested movie, using the obtained characteristics of each text and the corresponding label training support vector machine to obtain a sorting model of the texts; using the sorting model of the posters to predict the poster of the to-be-tested movie, to obtain a result Y1, and then calling the sorting model of the texts to predict the plot summary of the to-be-tested movie, to obtain a result Y2; and finally performing OR operation on the Y1 and Y2, to obtain the type of the to-be-tested movie. The method can rapidly and accurately predict the type of a movie under a condition of no movie videos.

A training supporting system has a training supporting apparatus and an exercise form analyzing apparatus, the both apparatuses being connected to each other through a communication network. The training supporting apparatus is worn on the arm of a user and measures acceleration rates of motion of the user's body where the apparatus is fitted on, at least in the three directions along X-, Y- and Z-axes, while the user is walking or running. Receiving the measured acceleration rates from the training supporting apparatus through the communication network, the exercise form analyzing apparatus analyzes an exercise form including balance between arm swing and foot landing, and sends back the analysis result of the exercise form to the training supporting apparatus through the communication network.

The invention belongs to the technical field of fitness equipment control, and discloses an embedded-based intelligent training support control method for a set of fitness equipment, the long-term observation and analysis of the physical health of the users using the sports fitness equipment are conducted, and students are guided to perform physical exercises scientifically according to the national standards so that physical fitness of the students can reach health standards; meanwhile, the accumulated data can be used for scientific researches of adolescent physical fitness and sports, health and disease and the like. The method allows consumers to enjoy top-notch personal trainer services without any additional costs so that ordinary consumers can acquire benefits and enjoyments from fitness. Coherent and complete fitness records are established for the consumers, so the consumers can visually see the changes of their physical fitness and enhance their confidence in fitness.