174 results about "Class model" patented technology

A system and method for performing object and class recognition that allows for wide changes of viewpoint and distance of objects is disclosed. The invention provides for choosing pose-invariant interest points of a three-dimensional (3D) image, and for computing pose-invariant feature descriptors of the image. The system and method also allows for the construction of three-dimensional (3D) object and class models from the pose-invariant interest points and feature descriptors of previously obtained scenes. Interest points and feature descriptors of a newly acquired scene may be compared to the object and/or class models to identify the presence of an object or member of the class in the new scene.

Audio/Visual data is classified into semantic classes such as News, Sports, Music video or the like by providing class models for each class and comparing input audio visual data to the models. The class models are generated by extracting feature vectors from training samples, and then subjecting the feature vectors to kernel discriminant analysis or principal component analysis to give discriminatory basis vectors. These vectors are then used to obtain further feature vector of much lower dimension than the original feature vectors, which may then be used directly as a class model, or used to train a Gaussian Mixture Model or the like. During classification of unknown input data, the same feature extraction and analysis steps are performed to obtain the low-dimensional feature vectors, which are then fed into the previously created class models to identify the data genre.

A system and method for recognizing instances of classes in a 2D image using 3D class models and for recognizing instances of objects in a 2D image using 3D class models. The invention provides a system and method for constructing a database of 3D class models comprising a collection of class parts, where each class part includes part appearance and part geometry. The invention also provides a system and method for matching portions of a 2D image to a 3D class model. The method comprises identifying image features in the 2D image; computing an aligning transformation between the class model and the image; and comparing, under the aligning transformation, class parts of the class model with the image features. The comparison uses both the part appearance and the part geometry.

Mental state of a person is classified in an automated manner by analysing natural speech of the person. A glottal waveform is extracted from a natural speech signal. Pre-determined parameters defining at least one diagnostic class of a class model are retrieved, the parameters determined from selected training glottal waveform features. The selected glottal waveform features are extracted from the signal. Current mental state of the person is classified by comparing extracted glottal waveform features with the parameters and class model. Feature extraction from a glottal waveform or other natural speech signal may involve determining spectral amplitudes of the signal, setting spectral amplitudes below a pre-defined threshold to zero and, for each of a plurality of sub bands, determining an area under the thresholded spectral amplitudes, and deriving signal feature parameters from the determined areas in accordance with a diagnostic class model.

A document classification system automatically sorts an input document into pre-determined document classes by matching the input document to class models. The content of the input documents changes with time and the class models deteriorate. Similarities between a training document set and an actual document set (which is classified into multiple classes) is calculated with respect to each class. A class with a low similarity is selected. Alternatively, classes where deterioration has occurred are detected by calculating similarities between the training document set in each individual class and the actual document set in all other classes. Class-pairs with low similarities are calculated. Close topic class-pairs are detected by calculating similarities between the training document set and all the class-pairs. Class-pairs with low similarities are selected.

The invention relates to a multiclass emotion analyzing method and a system facing a bilingual microblog text and belongs to the technical field of microblog text emotion analysis. The method comprises the following steps that (1) bilingual dictionary construction: corpus with an emotion inclination of a certain size is first collected, high frequent words with the emotion inclination can be extracted from the corpus, an emotional dictionary is then expanded by using an existing knowledge database and a vocabulary similarity calculating model, and finally network language and emotional signs are added in the emotional dictionary; (2) text pretreatment: the words are divided in a to-be-identified text, stop words are removed, and standardization treatment is conducted on English word shapes; (3) text characteristic space expression: the bilingual emotional dictionary is used for conducting vectorization on the text; (4) an emotional identifying task of the corpus text is realized through a multi emotion class model. The accurate rate and the F1 valve of the method are higher than those of a traditional classification method, and particularly the classification effect of a semi-supervised Gaussian mixture model classification algorithm in a small-scale training set is obviously better than that of the other methods.

The invention belongs to the technical field of personalized recommendation, and particularly relates to a method and system for solving a cold start problem in a collaborative filtering technology. The method for solving the cold start problem in the collaborative filtering technology comprises the steps that a data set is selected; an initial user or project clustering model is built through an optimized genetic algorithm; clustering is conducted on the basis of the initial user or project clustering model, and a user or project clustering model is obtained; entropy values of new users or new projects to all kinds of clusters in the clustering model are calculated, and the new users or the new projects are subjected to class cluster dividing; the new users or the new projects are recommended. The invention further provides a system for solving the cold start problem in the collaborative filtering technology. The system comprises a selection module, an initial model building module, a clustering module, a class cluster dividing module and a recommendation generation module. Accordingly, an improved genetic algorithm is utilized for conducting K-Means clustering, the initial user or project clustering model is generated, and recommendation is generated for the new users or the new projects.

a method for improving stacking schema for classification tasks, according to which predictive models are built, based on stacked-generalization meta-classifiers. Classifications are combined to build a new scheme from at least two layers and multiclass classification problems are converted into binary classification problems. One-against-all class binarization and regression learners are used for each class model and ensemble classifiers are improved using stacking. Accuracy differences, accuracy ratio, and runtime classification in multiclass datasets are also improved and the class of a value is then predicted.

A method of classifying samples to one of a number of predetermined classes involves using a number of class models or classifiers to form order statistic for each classifier. A linear combination of the order statistic (L-statistic) is calculated to determine the confidence of that particular classifier, both in general and for that particular sample. Relative weights are then derived from these confidences, and used to calculate a weighted summation across all classifiers for each class of the likelihoods that a sample belongs to that class. The sample is classified in the class which has the associated weighted summation which is greatest in value.

The invention discloses an intelligent steel cord conveyer belt defect identification method and an intelligent steel cord conveyer belt defect identification system. The identification method includes the following steps: (1) electromagnetic loading; (2) defect signal acquisition; (3) feature extraction; (4) training sample obtainment; (5) class priority determination; (6) multi-class model establishment; (7) multi-class model training; (8) real-time signal acquisition and synchronous class: electromagnetic detection units are adopted for real-time detection, detected signals are synchronously inputted into a data processor, features are extracted and then sent into established multi-class models, and the defect class of a detected conveyer belt is automatically outputted. The identification system comprises an electromagnetic loader, a plurality of electromagnetic detection units, the data processor and an upper computer, the data processor can automatically output the defect class of the detected conveyer belt, and the upper computer bidirectionally communicates with the data processor. The design of the invention is reasonable, the invention is easy to operate and convenient to put into practice, moreover, the using effect is good, the practical value is high, the reliability of conveyer belt defect detection is enhanced, and the efficiency of defect identification is increased.

A broadcast receiver includes a first video signal selector which outputs a video signal of a main-picture; a second video signal selector which outputs a video signal of a sub-picture; a first audio signal selector which outputs an audio signal of the main-picture; a second audio signal selector which outputs an audio signal of the sub-picture; a first video processor which generates a third video signal from the video signals of the main-picture and/or sub-picture; a feature quantity generator which generates a feature quantity of the audio signal of the sub-picture in accordance with a result of comparison between data corresponding to the audio signal of the sub-picture and sound class model data created beforehand; a determination processor which determines a scene as a specified scene; and a controller which controls the whole components in accordance with the specified scene.

The invention discloses a driving state recognition method based on approximate entropy template matching, and the method comprises the steps: 1, sample library building, wherein samples of one type in the sample library are a plurality of steering wheel turning angle signals in a normal driving state, and samples of the other type in the sample library are a plurality of steering wheel turning angle signals in a dangerous driving state; 2, road information segmentation based on the approximate entropy template matching: carrying out the call of a signal correction module based on the approximate entropy template matching to correct the steering wheel turning angle signals in the library sample, wherein the correction process of any one steering wheel turning angle signal is as follows: carrying out the EMD (Empirical Mode Decomposition), the effectiveness recognition of an intrinsic mode function component, and signal reconstruction; 3, feature extraction; 4, two-class model building and training; 4, driving state information collection and synchronous classification. The method is simple in steps, is reasonable in design, is easy and convenient to implement, is good in use effect, can accurately recognize the driving state of a driver simply and conveniently, and is high in recognition precision.

A method for generating application-specific input files for applications that require input files in varying file formats, where data for specifying industrial equipment are stored in a database of a computer system as XML-formatted files, which are based on a class model, and a modular transformation into at least one application-specific input file takes place. One or more of the XML files are loaded into a memory, decoded, interpreted and verified—particularly with regard to semantics and consistency. In the absence of an error specifications of objects of the industrial equipment are generated in XML format based on the verified XML files by using an interface containing descriptions of the functionality of the respective applications. The at least one application-specific file is generated via canonical transformation from the generated object specifications in XML presentation.

An embodiment of the invention discloses an intelligent interrogation system based on XGBoost disease prediction and a method. The system comprises a client and a service end which are connected through a network. The client comprises an interrogation module and a display module. The service end comprises a data processing module, a data storage module and an XGBoost disease predicting module. Through inputting patient symptom information which is acquired by a client into a trained XGBoost multi-class model, a disease probability predicted value and predicted disease information are output. Furthermore the patient can directly check the related information on the client. Based on machine learning and big data technology, a self-service diagnosis service of the patient can be realized. Thesystem and the method further have advantages of supporting intelligent analysis and prediction on mass patient data, realizing high accuracy and high timeliness in prediction, and realizing simple operation and low cost.

The present invention discloses a supply chain finance engine system, a system establishment method and a server. The system establishment method comprises the following steps of establishing a process definition file, which can be recognized and executed, by utilization of an on-line visual process design method; developing a business process execution engine, reading and analyzing the process definition file, and generating an execution class model; operating the execution class model, generating a process instance, executing process nodes, and generating a process task; and developing an on-line visual information presentation mode, and presenting the process operation conditions in a graphical mode during the process operation process. According to the supply chain finance engine system, participants such as core enterprises, upstream and downstream enterprises, banks and storage logistics service providers can be integrated, seamless combination with a supply chain business system is achieved, enterprise supply chain business activities can be organically combined with finance service activities, and huge income caused by supply chain management upgrade can be shared.

The invention relates to an evaluation method for performance influence degree of classification models by class imbalance. The evaluation method comprises the following steps of (1) building a classification model base; (2) constructing a new data set; (3) forecasting the new data set by the classification models; (4) evaluating the performance of the classification models; and (5) evaluating an influence degree level. According to the evaluation method, firstly, a typical classification algorithm in machine learning is adopted to build the classification model base; secondly, a class imbalance data set is selected as a reference data set, a group of new data sets with imbalance ratio gradually increased is built on the basis, different classification models are selected to respectively classify and forecast the group of new data sets; and finally, a variable coefficient is adopted to evaluate the performance variation degree of the classification models and also carry out level division, thus, the influence degree of the class imbalance on the performance of different classification models is evaluated, and a guidance significance is played in research on the class imbalance process. With regards to different classification models, the evaluation method for performance influence degree of the classification models by class imbalance, provided by the invention, has high universality.

The present invention discloses a method for accurately characterizing cut tobacco drying step material processing intensity. The method comprises: respectively collecting cut tobacco before and after drying; carrying out equilibrium; carrying out spectrum scanning; sequentially carrying out multiplicative scatter correction, second-order partial derivative-Norris derivative filtering, and main component analysis on the spectrum data; respectively establishing class models of the cut tobacco with no drying and the cut tobacco with different drying processing gradients; and calculating to obtain an average of a class model Mahalanobis distance from the cut tobacco with different drying processing gradients to the cut tobacco with no drying so as to determine a processing intensity. With the present invention, qualitative and quantitative expression of the cut tobacco drying intensity is achieved, objective guidance on parameter regulation and control of the drying step for preparing the cut tobacco thread is easily achieved, subjectivity influence on the existing processing intensity judgment is changed, and the intrinsic quality change of the cigarette can be well and objectively judged. In addition, the operation method is simple and easy to perform, and technical effects are remarkable.

The invention discloses a method for establishing a speech recognition model, a speech recognition method and a system. The method for establishing the speech recognition model comprises the steps ofreading and analyzing each grammar file and constructing a network of each grammar. Reading an acoustic syllable mapping relation table and expanding the network of each grammar into a syllable network respectively. The syllable networks are combined and minimized to form a phoneme decoding network; A speech recognition model is formed by using the phoneme decoding network and a language model. Byapplying the scheme of the invention, the expansion type is strong and supports N-Gram language model, supporting class model, flexible use, suitable for embedded recognizer in vehicle environment

The invention relates to a supervised learning method and device, a label prediction method and device, electronic equipment and a storage medium. The method comprises the steps of acquiring trainingsamples and a corresponding same-class reference set, wherein the same-class reference set comprises a plurality of reference samples, labels corresponding to the reference samples are the same as labels corresponding to the training samples, and the labels of the reference samples are noiseless clean labels; determining the weight of the training sample according to the similarity between the training sample and the same class reference set; and weighting the loss of the training sample by adopting the weight of the training sample, and updating the parameters of the to-be-trained classification model based on the weighted loss. According to the embodiment of the invention, computing resources can be saved, and the accuracy of the classification model can be improved.

The invention relates to a flood forecasting method based on a rainfall flood space-time database. The flood forecasting method comprises the following steps that 1) the rainfall flood space-time database is constructed; 2) a dynamic space-time rainfall map is drawn according to the rainfall flood space-time database with combination of GIS map data; 3) the space-time data corresponding to all the heavy rain paths involved in the flood in the historical periods of time are acquired according to the rainfall flood space-time database, and the heavy rain paths are displayed on the space-time rainfall map; and 4) corresponding class models are acquired according to the space-time data corresponding all the heavy rain paths, and flood forecasting is performed according to the class models. Compared with the methods in the prior art, the flood forecasting method has the advantages of being connected in space and time, dynamic and visual and accurate in forecasting.

The invention discloses a multi-version inference model deployment method, device and system in an edge computing environment. According to the method, an inference model instance deployment strategyon each edge computing node of the next time slot is continuously adjusted and updated according to the number of user inference requests arriving online, queues of to-be-processed inference requestsin each edge computing node at present and feedback of multi-version inference model instance deployment of the current time slot. According to the inference class model deployment system and deployment method, when the number of actual inference requests of a future user cannot be accurately obtained, by combining effect feedback of inference class model deployment after each deployment, periodicmulti-version model instance number adjustment is carried out on each edge node, and the effect of maximizing user inference precision in the edge environment is achieved.