315results about How to "Improve classification results" patented technology

A method for executing a transaction. A first processing device senses a sound, an action, or a behavior from a source and receives identification information from the source, which may be related to or a portion of the sound, action or behavior. The first processing device processes one or more of the sound, the action, and the behavior and the identification information to identify the transaction and to identify the source. Executing the transaction if the source is an authorized source.

The present invention relates to a method for the real-time identification of seizures in an Electroencephalogram (EEG) signal. The method provides for patient-independent seizure identification by use of a multi-patient trained generic Support Vector Machine (SVM) classifier. The SVM classifier is operates on a large feature vector combining features from a wide variety of signal processing and analysis techniques. The method operates sufficiently accurately to be suitable for use in a clinical environment. The method may also be combined with additional classifiers, such a Gaussian Mixture Model (GMM) classifier, for improved robustness, and one or more dynamic classifiers such as an SVM using sequential kernels for improved temporal analysis of the EEG signal.

The embodiment of the invention discloses a semantic segmentation method based on multi-scale convolutional neural networks (CNNs). The method includes: acquiring intra-modal features in high-resolution aerial-images and LiDAR (Light Detection And Ranging) point cloud data; carrying out inter-modal feature extraction and classification on the basis of the multi-scale convolutional neural networks;and using a multi-scale segmentation method to extract ground object boundaries, eliminate salt and pepper effects, and optimize classification results. By implementing the example of the invention,a method of combining the multi-scale CNNs and multi-scale segmentation (MRS) post-processing is used for semantic segmentation of the high-resolution aerial-images and the LiDAR point cloud data.

A method for operation of a search and recommendation engine via an internet website is described. The website operates on a server computer system and includes accepting text of a product review or a service review, initializing a set of words with seed words, predicting meanings of the words in the set of words based on confidence scores inferred from a graph and using the meanings of the words to make a recommendation for the product or the service that was a subject of the product review or the service review. The search and recommendation engine is also described.

The invention discloses a polarization SAR image classification method based on CNN and SVM, and mainly aims to solve the problem of the existing polarization SAR image classification method that the classification precision is low. The method comprises the steps as follows: (1) inputting a to-be-classified polarization SAR image after filtering; extracting and normalizing the original feature of each pixel point based on a polarization coherence matrix and by taking the neighborhood into consideration; training an AE network, and obtaining the parameter of a CNN convolution layer through softmax fine-tuning; setting a CNN pooling layer as average pooling, and determining the parameter of the CNN pooling layer; and sending the features of CNN learning to an SVM for classification to obtain the classification result of the polarization SAR image. Compared with the existing methods, the spatial correlation of the image is fully considered, a new neighborhood processing method is proposed based on CNN, features more conductive to polarization SAR image classification can be extracted, the classification accuracy is obviously improved, and the method can be used for polarization SAR image surface feature classification and object identification.

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.

The invention discloses a combination feature vector and deep learning based sentiment classification method and device. The combination feature vector and deep learning based sentiment classification method comprises the steps of obtaining multiple comment texts from internet; carrying out word segmentation for the multiple comment texts so as to obtain each sub-constituent words; obtaining a lexical feature of a sentence; extracting a syntactic feature of each comment text; obtaining a combination feature vector of each user comment text based on the lexical features and the syntactic features; and training a deep learning model based on the combination feature vector and further obtaining an optimal classification result through the deep learning model. According to the combination feature vector and deep learning based sentiment classification method and device, the comment text can be subjected to sentiment classification through the combination feature vector and the deep learning, the optimal classification result can be accordingly obtained, the classification precision is improved, and emotional tendency of a user in the text can be better identified; and the combination feature vector and deep learning based sentiment classification method is simple and convenient.

The invention relates to a combination-type text classification method based on deep learning. The method includes the following steps: preprocessing to-be-classified text; using the preprocessed textto obtain feature extraction results, which correspond to all deep learning models, through the different depth learning models; combining all the feature extraction results through a fully connectedlayer to obtain a combined feature vector; and using the combined feature vector as input to carry out classification recognition to obtain a class to which the text belongs. Compared with the priorart, the method has the advantages of lower requirements on training-corpus quality, high transplantability and expandability, a still better effect in a case of more classification classes and the like.

The invention relates to a method and a device for remote-sensed hyperspectral image classification. The method comprises the following steps of step 1: carrying out spectral angle weighted clustering based on kernel function vague C mean value on the hyperspectral image to obtain clustered indication characteristics; step 2: carrying out support vector machine (SVM) semi-supervised classification on the hyperspectral image to obtain a first classified image Image 1, and carrying out the SVM semi-supervised classification on the clustered indication characteristics to obtain a second calssified image Image 2; and step 3: establishing a clustering and SVM cooperation framework, inserting classification results of the Image 1 and the Image 2 into the clustering and SVM cooperation framework to be cooperatively analyzed so as to obtain a final hyperspectral classified image. The device comprises a clustering module, a classification module and a cooperative analysis module. The method and device for the hyperspectral image semi-supervised classification are feasible, capable of performing high-precise clustering and SVM-cooperative.

The invention discloses a polarization SAR image classification method based on RBM and SVM, mainly solving the problem of the existing polarization SAR image classification method that the classification precision is low. The method comprises the steps as follows: (1) inputting the polarization SAR image to be classified, having delicate polarization Lee filtering operation; (2) resolving and extracting original feature of each pixel point and integrating based on the polarization coherence matrix, polarization covariance matrix and Cloude; (3) initializing and training RBM to obtain the related parameter; (4) classifying by using SVM according to the feature learned by the RBM to obtain the classification result. Compared with the existing method, the space dependency of the image is totally considered and the feature in favour of classifying polarization SAR image can be extracted, the polarization SAR image classification precision is obviously raised, the method can be used for terrain classification and object identification for polarization SAR image.

The invention relates to a depth generating method based on a background difference sectional drawing and a sparse optical flow method. The depth generating method comprises the following steps: selecting suitable colour space and establishing a colour model; establishing a background model fast and effectively so as to exactly describe the background of a video sequence; classifying each pixel point according to the background model; optimizing the classified result of the pixel; evaluating an optical flow value of a feature point by the sparse optical flow method; evaluating the optical flow values of all the points by interpolation; and obtaining a depth value by the optical values. The invention has the advantages as follows: the sheltering problem which cannot be solved by singly using the optical flow method can be avoided; and the generation of the depth of the video sequence can be automatically realized.

The invention discloses a method for classifying polarimetric SAR (synthetic aperture radar) images on the basis of Cloude decomposition and K-wishart distribution and mainly solves the problem that the prior art is poor in classification effect and high in computation complexity. The method includes the steps of firstly, reading a polarimetric SAR image to be classified, subjecting each pixel of the image to Cloude decomposition to obtain an entropy H and a scattering angle alpha; secondly, initially partitioning the polarimetric SAR image according to the values of the entropy H and the scattering angle into eight classes; thirdly, subjecting results of eight classes of the whole polarimetric SAR image to iteration to obtain more accurate classification results. Compared with classic classification methods, the method is stricter in partitioning the polarimetric SAR image, better in classification effect, lower in computation complexity and applicable to terrain classification and target recognition of the polarimetric SAR images.

The invention relates to a remote sensing image ground object labeling method based on an attention mechanism convolution neural network, which comprises the following four steps: a computer reads data, constructs convolution neural network of attention mechanism, trains network model, and tests network to obtain labeling result. By adding an attention mechanism module, the invention enables the network to pertinently extract the information of the key position, makes up for the deficiency of the lack of the spatial information at the network end, and improves the classification effect of thenetwork to the ground object details. By using the mechanism of in-depth monitoring and using the characteristics extracted from the middle of the network to supervise the classification, the trainingspeed of the network can be further increased and the comprehensive performance of the network can be improved; through the up-sampling module of deconvolution, the resolution of feature extraction is increased and the method can overcome the problem that small objects are difficult detect to a certain extent, and can automatically classify remote sensing image pixels into corresponding object categories, reduce the trouble of manual interpretation, greatly accelerate the interpretation process, and obtain refined labeling results.

The invention discloses an image classification method and system based on migration learning. The method comprises the steps that 1, a feature similarity known training set A is utilized to make a training set B of a migration network through a support vector machine; 2, a migration learning network is constructed; 3, the training set B classified in the step 1 is used as a training learning setof the migration learning network, and a migration learning network model with high robustness and good accuracy is obtained through training; and 4, a to-be-classified dataset is introduced into thetrained migration learning network model, and a final classification result is obtained and marked with a tag. Through the image classification method and system, the requirement that a big sample dataset is needed to serve as input when a common RGB image is trained through deep learning is overcome, the problems of overfitting and locally optimal solutions in the training process are avoided, and classification precision is improved to some extent compared with a traditional classification algorithm.

The invention provides a geometrical feature-based human face aesthetics analyzing method, which comprises the steps of: carrying out combining description through combined strategy-based local geometrical features and constructing weak classifiers by using memory-based dynamically weighted kernel density estimation (MDKDE); and realizing effective integration of the features by using an Adaboost ensemble learning mechanism so as to obtain accurate classifications for human face aesthetics. Different from the traditional geometrical feature-based human face aesthetics analyzing technique, the geometrical feature-based human face aesthetics analyzing method selects the local geometrical features which are used for describing human face aesthetics from multiple angles, such as Euclidean distance, gradient and area and the like to compose single descriptions for the human face aesthetics, and combined feature description is obtained by combining descriptions of the local geometrical features, and the weak classifiers for Adaboost ensemble learning are constructed by using the MDKDE, thus a good classifying result for randomly input human face images is obtained.

The invention discloses a polarization synthetic aperture radar (SAR) image classification method based on spectral clustering. The polarization SAR image classification method mainly solves the problem that an existing non-supervision polarization SAR classification method is low in accuracy. The polarization SAR image classification method comprises the steps of extracting scattering entropy H of representation polarization SAR target characteristics to serve as an input characteristic space of a Mean Shift algorithm combining with space coordination information; diving in the characteristic space with the Mean Shift algorithm to obtain M areas; choosing representation points of all areas on the M areas to serve as spectral clustering input to spectrally divide all areas, and further finishing spectral clustering on all pixel points to obtain pre-classification results; and finally classifying the whole image obtained from the pre classification with a Wishart classifier capable of reflecting polarization SAR distribution characteristics in an iteration mode to obtain classification results. Tests show that the polarization SAR image classification method is good in image classification effect and can be applied to non-supervision classification on various polarization SAR images.

A method for detecting change in business data using a statistical classifier process. The method includes inputting a first set of business data in a first format from a real business process from a first data source and storing the first set of business data into one or more memories. The method also includes inputting a second set of business data in a second format from a real business process from a second data source and storing the second set of business data into one or more memories. The method forms a statistical classifier by inputting the first set of business data into a learning process associating with the statistical classifier that processes business the data in the first format. The method stores the classifier into the one or more memories, the classifier being associated with the first set of data in the first format and processes the data from the first data source in the statistical classifier to derive a first result. The method also processes the data from the second data source in the statistical classifier to derive a second result and determines a behavior of the statistical classifier based upon at least the first result and the second result. The method displays information associated with the behavior of the statistical classifier.

Techniques are disclosed for sound source localization based on joint learning and evaluation of ITD and ILD representations that are measured in a complementary, correlation-based way using binaural time-frequency spectrums. According to one embodiment, from these measurements and learned representatives, which may, for example, be created by combinations of measurements from signals belonging to the same class, i.e., the same azimuthal location, probability distributions over frequency and class are computed. These probability distributions can be combined over cue and frequency using information-theoretic approaches to get a robust classification of the location and additionally a confidence measure for the quality of the classification result.

The invention discloses a gesture recognition method based on a depth sensor. The method comprises the steps of sequentially obtaining depth flow information and human skeleton node data; cutting a gesture region; performing positive side and binaryzation are performed on 3D point cloud coordinates of the gesture region; gesture feature extraction, normalization and dimension reduction treatment are performed; gesture features are screened to obtain an optimal gesture feature subset; a support vector machine gesture training classifier is adopted, and a classification result is obtained; filtering is performed on the gesture classification result. The defect in the prior art is overcome, and the gesture recognition precision, stability and efficiency are improved.

The invention relates to an electromyographic signal tumble detection method based on a WKFDA. First, surface electromyographic signals are collected from the relevant muscle tissue of the lower limbs of a human body, the action signal segment of the surface electromyographic signals is determined through an energy threshold value, and fuzzy entropy is extracted from the surface electromyographic signals in the action signal segment to serve as characteristics to be classified; then, characteristic sample points are projected to a characteristic space, linear judgment is performed in the characteristic space, and therefore nonlinear discrimination of an original input space can be achieved in an implicit mode. Due to the contribution that corresponding balance weight is adopted for adjusting sample nuclear matrixes, influences of unbalanced data on the classification performance can be overcome. Due to the adoption of the nonlinear mapping, the data processing capacity of a Fisher linear discrimination algorithm based on nucleuses is greatly improved. The experiment result shows that a high tumble mode average recognition rate is achieved through the method, and the recognition result is superior to that of other classification methods.

An object-detection system includes a radar sensor, a camera, and a controller. The radar-sensor is suitable for mounting on a vehicle and is used to detect a radar-signal reflected by an object in a radar-field-of-view. The radar-signal is indicative of a range, range-rate, and a direction to the object relative to the vehicle. The camera is used to capture an image of a camera-field-of-view that overlaps the radar-field-of-view. The controller is in communication with the radar-sensor and the camera. The controller is configured to determine a range-map for the image based on the range and the direction of the radar detection, define a detection-zone in the image based on the range-map, and process only the detection-zone of the image to determine an identity of the object.

The invention discloses an unsupervised domain adaptation classification method based on inter-class matching, so as to solve the technical problem that the present classification method for images between different domains is poor in classification performance. According to the technical scheme, according to samples in a source image domain and a target image domain, the maximum average difference model for classes is built. A linear projection method is used to build a feature representation model for domain migration, the source domain samples and the target domain samples are projected to the same hidden feature space. In joint consideration of the supervision information of the source domain samples and hidden low-rank structural features between samples in the target domain, a robust target domain classification model is built, and all unlabeled samples in the target domain are marked. A joint optimization model with minimization of the distribution difference of the same class between domains as a target is built, an alternating minimization optimization method is used, alternating iteration of the feature representation model and the classification model is carried out until convergence, the optimal target domain classification result is obtained finally, and the classification performance is good.